THE

GEOLOGY OF AEEAK

IS LAND

A R RAN

f~\ Greenstone . basalt, , &p

THE

GEOLOGY OF ARRAN

THE OTHEE CLYDE ISLANDS

WITH AN ACCOUNT OF THE

BOTANY, NATURAL HISTORY, AND ANTIQUITIES

NOTICES OF THE SCENEKY AND AN ITINERARY
OF THE ROUTES

JAMES BRYCE, M.A., LL.D., F.G.SS. L. & I.

FOURTH EDITION

GLASGOW AND LONDON

WILLIAM COLLINS, SONS, AND COMPANY
1872

PREFACE TO THE FOURTH EDITION.

THE FIRST Edition of the following work was prepared by
request of the Local Committee of the British Association at
Glasgow, in order to serve as a guide to the Geologists attend-
ing the meeting in 1855. A sketch of the geological structure
of Clydesdale was the main feature of that edition. In the
SECOND Edition the portion treating of Clydesdale was much
extended, while that relating to Arran and the other Islands
was entirely new, and founded on observations of the Author
made during the intervening summers. It appeared to him,
however, that as Arran is a highly interesting field in other
departments as well as in Geology, the wishes of many
students would be met by combining an account of the
Botany and Marine Zoology with that of the Geology. At
his request the Rev. Dr. Miles, to whom a special grant had
been entrusted by the British Association for dredgings in
the Clyde, drew up an account of the more remarkable
objects of the Marine Fauna; to which was added a complete
list of the species then known, as well for the guidance of
students as a basis for future enquiry. An account of the
Flora of Arran was drawn up by the Author's eldest son,
Mr. James Bryce, now of Lincoln's Inn, and Regius Pro-
fessor of Civil Law at Oxford. In preparing the list
appended to it, advantage was taken of Professor Balfour's
kind permission to make use of his catalogue of Arran
plants. For this edition, which was published in 1859,
notices of the rarer Lepidopterous insects of Arran were most
kindly supplied by the distinguished entomologist, Mr.
Henry T. Stainton.

The THIRD Edition, published in 1865, was made much

2090849

VI PREFACE.

larger and more complete than its predecessor by adding
several new Excursions, by giving an account of the Arctic
shell beds of Arran, which had been in the meantime dis-
covered, and of the excavations within the Tormore Stone
Circles made by the late Duke of Hamilton, at the request
and under the direction of the Author, as well as by the
introduction of much new matter relating to the antiquities,
history, and traditions of the Island. The work was thus
made to serve as a general Guide to Arran. Dr. Miles being
absent from this country, several additional species were
added to the lists through the kindness of Dr. Carpenter.
Into the part relating to Clydesdale, changes not less extensive
were introduced; the more important of which was a con-
nected account of the Arctic shell beds of the Clyde basin, in
their true physical sequence, with a list of all the species
then known.

Since the Third Edition was published, the active and
successful labours of members of the Geological Society of
Glasgow have made the structure and fossils of the Clydes-
dale strata known to geologists everywhere, through the
medium of * their excellent Transactions, in much greater
detail than could be given in this work, and it has hence
been considered unnecessary to include in the present edition
that portion of the last which related to Clydesdale. The
.-juice thus saved has been used for the enlargement of the
rest of the work, so as to embrace more of the geology,
natural history, and antiquities of the Islands. A chapter
on the Birds of Arran has been most kindly supplied by .Mi.
Robert Gray, author of a recent work on the Minis of the
West of Scotland, which has already taken a high position;
and a list of the Quadrupeds has been obligingly furnished by
Mr. Edward Alston, F.Z.S., a well-known naturalist. The
Author has also to express his obligations to Dr. F. Buchanan
White, Dunkeld; Rev. David Landsborough, Kilmarnock;
and to Mr. David Robertson and Mr. Johu Young, Glasgow,
for important additions to the lists of species in the depart-

PREFACE. Vll

ments which they specially cultivate. The Author is indebted
to a relative, Mr. Robert Young, jun., for an illustrated de-
scription of the antiquities of Bute, of which no connected
account has before appeared; and to the skilful pencil of Mr.
Young's father for most of the illustrations of the present
edition. Mr. Stephen Miller has spared no pains to render
these faithfully in his lignographs, which it is hoped will add
to the interest and value of the work.

GLASGOW, June, 1872.

CONTENTS.

GEOLOGY OF ARRAN.

PACK

PHYSICAL FEATURES —

Dimensions — Interest of the Island — Mountain nucleus —

Hill Ranges — Heights — Southern plateau, ... 1

OUTLINE OF GEOLOGY —

Granite abnormal in position — The slates — Granites of the
nucleus, coarse and fine — Old red sandstone — Carbon-
iferous series, 6

THE OUTLYING GRANITES —

The Ploverfield granite — Granite of Craig-Dhu — Age of
the granites, 16

THE TEAPPEAN ROCKS —

Their limits, and mineral distinctions — Dikes, their number

and range, .28

GLACIAL ACTION —

Evidences of it — Terraces and mounds — Dispersed blocks
Theories of Glacial action — Arctic shell beds, . . 35

EXCURSIONS IN ARRAN.
I. To THE SUMMIT OF CIOR-MHOR —

Its advantages as a first excursion — The Brodick monolith
— Section of strata — Glen Rosa granite junction — The
Garbh-Alt — Grand scenery — Dikes — Mountain plants —
Origin of the Ceims — Summit of Cior-Mhor — Effects of
light, . . . ' . «" . . , . . 49

II. TO THE CORRIEGILLS SHORE —

The coast sandstone — Their dikes and beds — The great
boulder — Dun-fion range and fort — Porphyry and pitch-
stones— Evening aspects, 68

X CONTENTS.

III. To THE SUMMIT OF GOATFELL — I>A(;K
Meaning of the name — Rocks on the ascent — Chasm on

dike — Junction at mill-dam, and singular dike —
Panorama from summit —Cyclopean walls and N.
Goatfell — Origin of the Corries — Plateau S.W. of
Goatfell and roche moutonnee, ..... 84

IV. THE WAYSIDE MUSEUM —

Its locality and specimens —The old rocks— Igneous
products —Their mineral structure and mutual rela-
tions—Origin of the columnar structure — The Birk
Glen — The Arran burns, 93

V. BY THE CORRIE SHORE—

View from Invercloy — Parochial divisions of Arran —
Brodick Castle — The Hamilton family — Sandstone
and dikes— Lily's cave — List of shells — Rocks of
Maoldon— Junction in White Water, . . . 100

VI. To THE NORTH SHORE —

Corrie-The base of two systems — The anticlinal axis
— Sandstones, limestones— Fossils of the trap]>can
ash — The coal beds— Singular section at Newton-
point —Loch I Ian/a and its Castle— History— Scenery, 1 1 8

VII. BY THK WESTERN SHORE—

Glen Catacol— Anticlinal in the slates— Terraces and
mounds — Slates of the coast— lorsa water -foot —Age
of the terraces, 1 35

VIII. BY THE INTERIOR OF THE NUCLEUS—

Gorge and ancient glacier — Torneadaneoin junction —

The two granites— Dikes in the fine, . . . 140

IX. LOCH RANZA TO CORRIE—

Glen Chalmadale junction — Glacial striae— A desertc<l
village, 145

X. To HOLY ISLE—

Structure of the mountain— The east shore— Correct
legend of St. Molios— Kilbride Church, . . . 148

XI. To WINDMILL HILL AND PLOVERFIELD —

Kilmichael — The Fullarton family— Porphyry, sand-
stone, granite, and syenite — Amethysts in sandstone —
Ben-leister Glen— The pitchstones of Moneadh-Mhor
Glen, . * 152

CONTENTS. XI

XII. To CEIM-NA-CAILLIACH AND THE CASTLES—

Crystals of smoke-quartz — Analysis of felspar— The PA«E
Carlin's Step— Castle Abhael, ..... 159

XIII. GLEN SANNOX AND GLEN ROSA —

Ancient Chapel — Barytes veins — Discovery of the pass —
Great dikes — Singular pitchstone dike — The adder,
dissemination of species — Meteorological phenomena —
The Arran Sabbath, 162

XIV. BY THE EASTERN SHORE TO DJPPEN AND LAG —
Lamlash Bay — Final Causes — King's Cross — Glen Eais-

dale — Dippen cliffs — A basaltic causeway— Kildonan
Castle — Eais-a-Mhor — Struey rocks and Benan-head —
Lag, 169

XV. LAG TO THE SHELI, BEDS —

Lag — Fault and dikes — The Arctic shell-beds— Sections
and causes of error — Beds on Cloinid burn — On Torlin
Water — On Slaodridh Water — Crook-crever burn —
General Geology — Cairn and fort, . . . .180

XVI. To DRUMADOON AND KING'S COVE—

The String road and watershed — Craig- Dhu — Clay-
stones of Black water-foot — Drumadoon and King's
Cove — The Tormore pitchstones — Analyses — The
Leac-a-Bhreac porphyry— Clachan Glen, . . . 193

ANCIENT REMAINS OF ARRAN —

General Notices, 206

Excavations at the Stone Circles, 209

FOSSILS OF THE ARRAN ROCKS —

The Slates, . . . . . . . . .223

The Old Red Sandstone, 223

The Carboniferous Series, ...... 224

THE FLORA OF ARRAN —

Relations of the Flora to that of the adjoining tracts, . 228
List of Plants, 239

MARINE ZOOLOGY OF ARRAN —

Notices of characters and habits, 252

List of Species, 276

Xll CONTENTS.

ENTOMOLOGY OF ARRAX — '*<>*

Notices of rare species, 285

THE BIRDS OF ARRAX, 288

List of Species, 292

THE MAMMAL-; OF ARRAX, 313

GEOLOGY OF BUTE, . 316

THE ANTIQUITIES OF BUTE, . . 340

THE OTHER CLYDE ISLANDS —

Inohinarnock, 349

Great Cumbrae, .349

Little Cumbrae, ........ 351

Pladda, 352

Ailsa Craig, . . . . .... . 353

GEOLOGY OF ARRAN.

1. THE Island of Arran is twenty miles and a half long from
N.N.W. to S.S.E. ; ten and a half miles broad, and with
Holy Isle and Pladda includes an area of 103,953 acres. —
The number of rock formations, sedimentary and plutonic,
which are found within this limited space is truly remarkable,
perhaps unparalleled in any tract of like extent on the surface
of the globe; while the varied phenomena which they present
in their mutual contacts and general relations to one another
are of the highest import in theoretical geology. The variety
indeed is so great, and the interest so lively and pleasing,
which an examination of the structure of the island and its
charming scenery excite, that, as Professor Phillips has re-
marked, every geologist who visits Arran is tempted to write
about it, and finds something to add to what has been already
put-on record. For the student there cannot be a finer field;
the primary azoic rocks, the metamorphic slates, the lower
palaeozoic strata, the newer erupted rocks, and phenomena of
glacial action, may all be examined by him in easy excursions
of a few days; and the exposition of the strata is so complete
in the rugged mountains, deep precipitous glens, and unbroken
sea coast sections, that the island may truly be called a grand
museum, arranged for his instruction by the hand of Nature.

Physical Features.

2. A line running from the north angle of Brodick Bay,
almost due west to lorsa water-foot, divides the island into
two nearly equal poi-tions, strikingly different in their geolo-
gical structure and in their outward features. The northern

B

2 GEOLOGY OF ARRAN.

half consists of a mass of peaked and rugged mountains, in-
tersected by deep and wild glens, which diverge from a
common centre, and open seaward on a narrow belt of low
land. This belt forms a terrace marking the ancient sea-
level, and is bounded inland by cliffs pierced with caves,

Fig. 1.

and otherwise sea-worn. The coast road is carried along it
from Brodick Bay to the mouth of South Sannox Water, and
again, from Loch Ranza to Dougrie, at the mouth of lorsa
Water, and affords throughout views of surpassing beauty.
This terrace is a striking feature of the Clyde shores every-
where, and will be more fully described farther on.

The Ai'ran mountains are naturally divided into three
separate Ranges, which may be named after their most con-
spicuous summits — the Cior-Mhor (Kior-Vawr), Goatfell,
and Ben-Bharrain (Varen) groups. Of these, by far' the
most considerable is the first, which forms a long, irregular,
narrow, and jagged ridge, extending from Ben-Ghnuis (Nuish)
on the south to Suidhe-Fergus (swi-C-) on the north-east.
Cior-Mhor stands near its centre, at a point where a salient
angle in the ridge closely approaches to the Goatfell group
on the east. Its connection to this group by a cross ridge or
col, 1250 feet in height, which separates the heads of Glen
Rosa and Glen Sannox, and its position near the middle of
the range, constitute the prominent peak of Cior-Mhor the
geographical centre of the whole north-eastern mass of
mountains. The western Range has great persistent altitude,
no point descending below 1GOO feet, and there being at
least six summits not less than 2000 feet. Fronting the

PHYSICAL FEATURES. 3

concavity of this arch-like ridge on the east, is an assem-
blage of closely-connected mountains dependent upon Goat-
fell. The most northerly summit is the lofty conical peak of
Cioch-na-h'oighe (Kioch-na-oiche), or the Maiden's Breast,
guarding the south-east angle of Glen Saunox; on the south
the group terminates in the bold precipice called Gleu-
Shant rock, at the entrance of Glen Rosa. Both groups
front the interior glens in tremendous precipices, while they
descend with less abruptness to Glen lorsa on one side, and
the sea on the other. To the west of the geographical axis
of the island lies the Ben-Bharrain range, with some con-
nected and lower heights east of it, the entire group being
separated from the other mountains by Glen lorsa and Glen
Eas-an-Bhiorach (e-as), running respectively S.S.W. and
N.N.W. through the length of this half of the island. These
glens have a common watershed in Loch-an-Davie, a small
mountain lake or tarn, which, when it stands at a high level,
as in winter and in wet summers, discharges its waters at both
ends. The volume of water discharged into Glen lorsa is
indeed greater than that flowing out into the north glen; yet
the origin of this latter, and its outlines towards the top, are,
as plainly as in the case of lorsa, due to the action of the
little stream — it is, in fact, towards the lake as a centre that
the slopes of both glens manifestly converge. The existence
of the double discharge has indeed been called in question.
We have, however, observed it ourselves on several visits,
repeated after a considerable interval, in the months ranging
from April till October ; and it is now laid down as a recog-
nized feature1 of the district on the maps of the Ordnance
Survey, published since the issue of our previous edition.
This singular geographical arrangement, said to be exhibited
on a grand scale by some lakes in Canada and Norway, is
repeated to the south-west of Ben-Bhreac (vre-ac), the high hill
which overhangs the west side of Loch-an-Davie. Here a rill,
coming down from the hill-side, on reaching a wide hollow,
which connects a branch of Glen Catacol to a branch of Glen

GEOLOGY OF ARRAN.

lorsa, is seen to part into two streamlets, half the water
running towards Catacol, and half towards lorsa. A shallow
pool or small lake sometimes occupies this hollow, and there
is then a double outflow, as in the case of Loch-an-Davie.*

The bleak uplands between Ben-Ghnuis aDd Ben-Bhar-
raiu are finely varied by several heights, of which the most
remarkable is the prominent ridge of Sail-Chalmadale (sa-Il),
an outlier of the latter range. It divides Glen lorsa from Glen
Scaftigill, and is a conspicuous and picturesque object from all
the south-western portions of the island. — To complete our
sketch of the northern section, it is only necessary to notice
the high ridge between the sea on the north, and north Glen
Sannox .and Glen Chalmadale on the south. In geological
formation it differs from the others, but is connected with
the main range, which we have called the Cior-Mhor group,
by means of a ridge at the watershed between the two glens
which bound it. This ridge joins on to the eastern flanks of
Torneadaneoin (Ne-ad-an-o-in), an abrupt, massive, but
not lofty mountain, overlooking Loch Ranza on the south,
and forming the termination of a long but not generally high
ridge which sweeps round the east side of Glen Eas-an-
Bhiorach, and runs in upon the north side of Caistael-Abhael
(ca-Ts-tel-avel), near the watershed at Loch-an-Davie. This
northern range terminates in bold heights forming the east
side of Loch Ranza; its western boundary is formed by the
nearly precipitous sides of Meall-Mhor (me-al), the most
northerly mountain of the Ben-Bharrain group. Thus, at
the northern apex of the island, the principal ranges closely
approach one another, their terminal portions forming the
lofty abrupt framework to the secluded inlet of Loch Ranza,

* A ditch at the summit level of the String road, between Brodick
and Shiskin, has a like outflow to both seas. The same structure is
seen on the road* leading from Dingwall towards Skye ; nearly two
miles west of Auch-na-Sheen a spring gushes out from the middle of
the road, and running into the adjoining ditch parts off both ways,
half to the Atlantic, and half to the German Ocean.

PHYSICAL FEATURES. 5

whose many picturesque features excite the admiration of
everv visitor.

The heights of the Arran mountains have lately been
determined, in the conduct of the Ordnance Survey, with that
exactness which marks all the operations of the Royal
Engineers. The following are the heights of the principal
points: — Goatfell, 2866 feet; Caistael-Abhael, 2735; Ben-
Tarsuin, 2706 ; north top of Goatfell, 2628; Cior-Mhor, 2618;
Ben-Ghnuis, 2597; Ben-Bharrain, south summit, 2345 ; north
summit, 2333 ; Cioch-na-h'oighe, 2168 ; Am Binnein (over
High Come), 2172 ; Ben-Cliabhain (W. of Glen. Rosa),
2141 ; Ben-Bhreac (W. of Loch-an-Davie), 1881 ; Loch-
an-Davie, 1182; Meall-Mhor, 1602; Meal-nan-Damh (W.
of Glen Catacol),1870; Sail-Chalmadale, 1572; Loch Tanna,
1065; Loch Ghnuis, 1069; Loch lorsa, 146; Maoldon,
1206; Torneadaneoin, 1057; An Tanna (K of String
road), 1184 ; highest point of the slate ridge, E. of Loch
Ranza, 1453.

3. The southern half of the island consists of a rolling
table-land, bleak and unpicturesque inland, but breaking
rapidly down seaward into a coast border of great romantic
beauty. The general elevation is from 500 to 800 feet ; and
the irregular ridges which traverse it, most usually in a
direction nearly east and west, rise from 1000 to upwards
of 1600 feet. The highest points are: — Ard Bheinn,
1676 feet ; Cnoc Dhu, north branch of Glen Cloy,
1341; A'Chruach, head of south branch, 1679; Cnoc-na-
Croise, head of Clachan and Ben-leister glens, 1346 ; Beinn-
Bhreac, north-west of this, 1649; Tigh-vein, S.W. of Urie
loch, highest point south of the parallel of Lamlash, 1497;
other heights west of this, 1385, 1367 ; west of the high
road, head of Sliddery Water, Cnoc-na-Chapuil], 1369; Cnoc-
na-Dail, 1158; farther west, Tormusk, 1259; Cnoc-an-Loch,
1107; west of Whiting Bay, Cnoc-Mhor, 867, Borrach, 869;
Cnoc-an-Fheidh (he-g), 875; Cuoc-an-Garbad, east of Loch
Garbad, 959 ; Auchinhew, 887 ; Holy Isle, 1030 ; Bennan

b GEOLOGY OF ARRAK.

Head, 457 ; West Bennan, 523 ; highest point near the south-
west coast on the muir of Corricravie, Cnoc Reamhar, 737
feet. The interior heights separate the various glens and
river-courses, whose origins lie near the central line or
axis of the island, but often interlace with one another,
so that a stream issuing westward has its source nearer
the east coast than the west, and vice versa. The views
of the northern mountains from these uplands are very
grand, especially when they are seen in the early sum-
mer twilight, their dark jagged peaks projected against
a background of sky, still lit up with brilliant hues from the
departed sun. The terrace border, so conspicuous around
the estuary of the Clyde, is marked in the southern section
of the island with less continuity than round the northern
part, owing to the nature of the rocks, and their advance in
many places upon the sea-line in mural precipices. Still,
however, it is sufficiently distinct in many parts, as about
Corriegills, Whiting Bay, portions of the south coast, and
towards King's Cove on the west, where its salient and re-
entrant angles, in their bold points and noble sweeps, enclose
some of the finest scenes of quiet beauty to be met with in
Scotland.

General Outline of the Structure.

4. The remarkable geological structure of Arran, and the
striking physical features which give such a charm to its
scenery, are alike due to a single peculiarity — the abnormal
position of its granite nucleus. Granite usually forms a
middle dominant ridge, or mineral axis, to which the lateral
ridges on either side are subordinate and equally related, as
in the annexed ideal section, fig. 2, where a is the granite
axis ; bb, old slates corresponding on opposite sides ; cc,
younger slates, and so on. In Arran, on the other hand,
the granite does not form a mineral axis, but has been
protruded close to the outer border of the two upper slates, so
as almost to come in contact on the E. side with the newer

GENERAL OUTLINE OF THE STRUCTURE. 7

sedimentary strata. These and the slates are not repeated
on the W. side in like positions, the strata being successively
older as we advance in that direction. The slates thus

«, Mineral axis of granite; b fej old slates corresponding on opposite
sides of the axis; c c and d d newer slates also corresponding.

pierced are part of a broad zone of slate rock which traverses
Scotland diagonally from sea to sea, intersects the line of the
Clyde shores, and crosses the islands of Bute and Arran. It is
usually divisible into thi'ee distinct bands; a lower micaceous,
passing into gneiss; a middle dark-coloured clay slate, and an
upper green or chlorite slate, the two latter forming what is
called the clay-slate series. The newest of the two upper slates
is not found in Arran, and is probably either thrown out west-
ward beyond the line of bearing of the second slate, or is so
altered by the near proximity of the granite as to be undis-
tinguishable from the middle or dark-coloured slate, through
which the granite has been protruded. On the east side of
the granite nucleus, above Corrie, this slate band is extremely
narrow ; on the western side it is much broader ; but the
lower micaceous band appears only in patches in the promon-
tories west of Catacol ; it is not seen in any great body till
we pass into Cantire.

This singular abnormal position of the Arran granite,
first pointed out by Dr. M'Culloch, gives to the island all its
peculiarities, both as regards its geographical features and its
geological structure. The protrusion of so large a body of
igneous rock along the line of junction of the older slates
and the secondary formations, and its elevation to a great
height in a space so limited, might naturally be expected to
produce phenomena of varied interest, such as have been
alluded to in the opening paragraph (Art. 1).

GEOLOGY OF ARRAN.

THE SLATES. 9

The Slates.

5. The granite nucleus occupies by far the greater portion
of the northern half of the island. The three mountain
groups already described, with the glens and valleys pene-
trating and dividing them, consist entirely of this rock. It
is remarkable, however, that at no point does the granite
reach the sea coast. It is everywhere enclosed by a narrow
band or framewoi'k of clay slate, of the second or dark
coloured variety, which completely encircles the nucleus
(see Map). The structure of the northern half of the island
is shewn in the preceding section (fig. 3), which crosses the
nucleus east and west from Corrie to White Farland. In this
section, a is coarse-granite ; b, fine granite ; c, slate ; d, old red
sandstone; e, sandstone with coal plants;/, limestone with
product!, corals, <fec. ; g, dikes of basalt and pitchstone.

From the mouth of the lorsa Water at Dougrie, round
the west side of the island, by Imochar, ;Thundergay,
and Catacol, to Loch Ranza, the slate rock occupies
the coast, and forms a belt of considerable but varying
breadth. Its junction with the granite is seen in almost
every stream, and in many points along the western slopes
of the lower hills. It extends all round the precipitous
sides of the Loch Ranza valley, and to a short distance
east of Newton point, which forms the north-east angle of
the loch, as far as a small stream called Alt-Mhor. It here
retires from the shore, and forms the high northern ridge
already mentioned; the coast from this point eastwards, and
then southwards, being occupied partly by old red sandstone,
and partly by coal sandstones, beds of carboniferous lime-
stone and of coal and coal shale being interposed amid the
beds of the latter. The band of slate is of considerable
breadth in this northern ridge, but narrows very much on
approaching the opening of Glen Saunox ; along the hill-sides
southward, from the base of Cioch-na-h'oighe to the slopes
over Con-ie, the breadth vai'ies from 25 to 40 or 50 yards.
Farther south, as it sweeps round south-west between

10 GEOLOGY OF ARRAN.

Maoldon and the base of Goatfell, it gradually widens —
presents bold precipices on both sides of the lower part of
Glen Rosa — and along the south border of the granite nucleus,
between Loch Ghnuis, Dougrie, and Imochar, attains its
greatest development. Here the thickness of the slate band
between the granite and old red sandstone cannot be less
than 2000 or 3000 feet; whereas above Corrie it forms but
a thin band. The granite here closely approaches the old
red sandstone, but is nowhere in contact with it. The pre-
vailing dip of the slate over the whole district is the same
as that which it maintains in adjoining tracts beyond the
limits of Arrau — namely, south or south-east at a high angle;
it is not arranged in mantle-shaped strata around the granite
nucleus, but is inclined towards it in some places, and off
from it in others. Thus, along the north side of the
nucleus, from North Sannox to Catacol, the dip is south or
towards the granite, at angles varying from 65° to 75°. This
is of course unconforuiable to the granite centre ; but a like
dip and inclination on the south side give conformability ;
and here, accordingly, from Maoldon by Glen Rosa and
Dougrie, and perhaps even as far as Thundergay, the slate
is seen to recline against the sides of the granite mountains
in a kind of mantling stratification. In many spots there is
great contortion and irregularity, indicating the operation of
violent forces, attendant on the upheaval of the granite, and
in several places, as above Corrie, and near Catacol, the
strata come against the granite end on, or nearly at right
angles. Reference will again be made to several of these
cases, as well as to local variations in the mineral character
of the slate.

Granites of Hie Nucleus.

G. The coarse and fine granites, as above traced, are found,
wherever they co-exist, in the same relative position ; the
coarse underlies the fine-grained. The outward aspect and
structure of the two kinds are very distinct, though the com-

GRANITES OF THE NUCLEUS. 11

ponent minerals are the same. Both consist of quartz, fel-
spar, and mica, the two first being in nearly equal proportions,
the last in less quantity than either. In neither kind is the
mica replaced by hornblende, so as to form the variety called
syenite. This rock is indeed found in Arran, but not within
the district called the granite nucleus. There are, however,
several varieties of both kinds; but these are merely depend-
ent upon slight changes in the colour or size of the constituent
crystalline grains. The fine variety has often an arenaceous
aspect and sandy feel, from the minuteness of the grains. It
has a different " ring" under the hammer, and affects the
form of huge flat slabs or sheets, rather than the massive
rhombic-block form of the coarse granite, so well seen in the
" Cyclopean walls" of Goatfell, Ben-Ghnuis, and Cior-Mhor.
The distribution of the two granites within the area of
the nucleus is shewn on the accompanying map. The over-
lying fine-grained granite is by no means co-extensive with the
coarse-grained ; it is, in fact, almost limited to the mountain
sides and lower hill-tops within the lorsa basin; only a small
area of it spreads out across the watershed towards the north.
The coarse-grained variety constitutes the entire mass of the
Goatfell group, from the mill-dam round the convexity of the
great arch as far as Cior-Mhor, the bottom and sides of Glen
Rosa, the east front and highest crests of Ben-Ghnuis and
Cior-Mhor, the ridge of the Ceims, which connects them, Glen
Sannox, and the southern fronts and summit ridges of Cais-
tael-Abhael and Suidhe-Fergus. It occupies also the bottom
of Glen lorsa and the sides of that glen, to the height of
several hundred feet, running out south to contact with the
slate along the hill-sides between Loch lorsa and Loch
Ghnuis, and extending up the glen northwards as far as the
great waterfall, about a mile south of Loch-an-Davie. This
same variety forms the central and lower part of Glen Eas-
an-Bhiorach, the ridge on its east side, and the mountains on
the east side of Glen Catacol. The fine-grained variety
occupies the plateau of Loch-an-Davie, and extends some dis-

12

GEOLOGY OF ARRAN.

tance northwards, on both sides of the stream, down into
Eas-an-Bhiorach. On the western side of the lorsa valley
it is continuous over the whole surface, from the level of the
great fall on the lorsa to the summits of the hills, and attain-
ing, on the east side of Ben-Bharrain, an elevation nearly as
great as the depression between the two summits, which, as
well as the whole central and western parts of this group,
are composed entirely of the coarse variety. The junction
of the two kinds and their interlacing veins are finely seen
on the southern front of Sail-Chalmadale.

On the eastern side of the lorsa valley the fine-grained
granite stretches continuously along the western base of the
Cior-Mhor and Ben-Ghnuis group, rising high on the western
side of these mountains, Caistael-Abhael and the Ceims, and
forming the summits of the passes into Glen Sannox and
Glen Rosa, on the northern and southern sides of Cior-Mhor.
Mantling round the southern flanks of Ben-Ghnuis, the fine
granite runs out against the slate, which it penetrates and
alters. The relation of the rocks here is shewn in the an-
nexed diagram (fig. 4), which is a ground plan : —

c . Fig. 4. c

a, Coarse granite ; b, fine granite ; c, slate altered and banded.

The coarse variety is seen in close proximity, impenetrated
also by veins of the fine-grained granite. By the fall of the
ground both ways — eastwards towards the Garbh-Alt and
Glen Rosa, and westwards towards the lorsa — the coarse-
grained granite soon emerges from beneath the fine, and thus

THE OLD RED SANDSTONE. 13

the contact of the latter with the slate along this, their only
line of junction, is of very limited extent. At no other part
of the boundary does the fine granite meet the slate band;
and here it alters the slate in the same manner as the coarse
granite : the slate is, in fact, converted into a banded rock,
in which a partial separation of the ingredients has been
effected, and a variety produced closely resembling a dark-
coloured gneiss. The annexed section (fig. 5) will best shew
the relation of the rocks at this interesting junction : —

Fig. 5.

c, Slate penetrated and altered by fine granite ; b, fine granite ;
a, coarse granite, forming the upper slope and top of Ben-Ghnuis ;
d, rock uncertain.

The Old Red Sandstone.

7. The encircling band of clay slate is succeeded on the
east and south by a band of old red sandstone, which, like
the slate band, is of irregular breadth. It begins to overlie
the slate at the Fallen Rocks on the north-east coast, and
occupies the shore thence to the march of Achab farm, half a
mile north of Corrie. Here it retires inland, the carbon-
iferous formations taking its place on the shore; crosses in a
narrow band to the west of Maoldon, and stretches thence
continuously westward, around the border of the slate, to the
mouth of Mauchrie Water. Between this point and Dougrie,
near lorsa water-foot, it attains its greatest breadth. The

GEOLOGY OF ARRAN.

breadth is also considerable from the Fallen Rocks to a point
a little south of the base of Cioch-na-h'oighe. A line from the
north side of Brodick Bay to Dougrie very nearly marks out
its line of junction with the slate. In structure it varies from
a fine-grained red or dark-brown sandstone to a coarse con-
glomerate, in which the fragments are more than a foot in
diameter. The coarse and fine strata do not follow any
particular oi-der, but alternate throughout the formation,
indicating the operation of powerful currents and interme-
diate periods of repose.

It may here be merely noticed that the sandstones of the
coast have the edges of the strata tm-ned up towards the
central granite, and that north of the mouth of North San-
nox Water the dip is towards the north-east and north,
while south of this point it is south-east and south. Thin
bands of limestone occur subordinate to the old red sandstone
at the march of Achab farm at Corrie, in Glenloig, and in a
few other places. This limestone is of concretionary struc-
ture, without fossils, and similar to the cornstoues of Eng-
land— members of the Old-Red system.

T/te Carboniferous Series.

8. The southern half of the island, southward of the old
red sandstone boundary above defined — that is, from the
String road and valley of Mauchrie Water to the South End —
is composed of several members of the carboniferous series,
broken through and overlaid by various igneous rocks, chiefly
those of the greenstone type. The prevailing rock and sub-
stratum of the whole southern plateau is red sandstone, vary-
ing from a fine compact structure to that of a coarse con-
glomerate. A band of this sandstone extends also along
the eastern shore northwards to Corrie ; and again from
the Fallen Rocks north-west to the Scriden at the northern
extremity near the entrance of Loch Ranza. Subordinate to
this sandstone are beds of limestone, abounding in fossils of
tine carboniferous types, with beds of shale and coal, in which

THE CARBONIFEROUS SERIES. 15

are found fossil plants and shells, such as characterize these
strata in the basin of the Clyde. The structure, indeed, is
quite analogous to that of the Clyde basin. Limestone does
not occur at the base of the system, nor does it occupy any
determinate place in it, but is found throughout the whole
series of beds in repeated alternation with the sandstones and
shales. These alternations are seen in section on the sea
shore along the east border, on the hill-side between Corrie
and Brodick, along the high grounds on both sides of Glen
Cloy, in Clachan Glen, Glen Scorodale, and other water
channels issuing westward ; and in the Ben-leister burn above
Lamlash they descend almost to the bottom of the glen. On
this ground we refer the whole of the sandstones of the
southern division of the island to the carboniferous series.
The alternations in question are not seen, it is true, on the
Corriegills, Whiting Bay, or Kildonan shores, nor do they
appear among the sandstones of the South End ; but such
massive strata of red sandstone as appear along the east and
south shores are common in the carboniferous system, and
form indeed everywhere its prevailing member. They are
conformable to the strata inland, in perfect sequence with
them, of the same mineral structure, and therefore referable
no doubt to the carboniferous system. They may thus be
correlated with the calciferous sandstones of the Ayrshire
coast, upper members of the system, described by the geolo-
gists of the Ordnance Survey ; but this can only be regarded
as a provisional arrangement until the discovery of fossils
shall give us more certain data for classification.

9. The fragments imbedded in the conglomerate, both of
this age and of the Old-Red system, are mica slate from a
distant source, the ordinary slate adjoining, and quartz of
two varieties, of which one is the common white quartz,
forming veins in the slate, and the other a peculiar resinous
quartz, of a cinnamon colour, not found in Arran, but closely
resembling certain varieties found on the west coast, and in
Islay and Jura. The resinous variety is rounded and polished,

16 GEOLOGY OF ARRAN.

a distant source being thus indicated; while the other is
often angular or slightly rounded. It is remarkable that frag-
ments of granite do not occur in the conglomerate ; none of
the Arran sandstones have as yet yielded a piece of this rock,
except in a single instance, where its presence may be other-
wise explained (see Art. 12). The conclusion to be drawn
manifestly is — that when these sandstones and conglomerates
were in process of formation by the wearing down of the
slate rocks and the transport of the fragments by water, the
granite of the interior was not exposed to disintegrating
causes, but remained as yet in hypogene depths, protected
most probably by the enveloping slate rocks. Facts to be
stated farther on will throw light on this curious subject.

The, Outlying Granites.

10. Perhaps the most remarkable feature in the geology
of Arran is one made known by its recent explorers, and
of which the British Isles offer, we believe, only one other
example. We refer to two outbursts of granite amid the
sandstones of the southern division of the island, noticed in
the last three Articles. One of these was discovered by Mr.
Ramsay in 1837; but the first description of it was pub-
lished in 1839 by M. Necker, who named the district
Ploverfield. The other was discovered by the writer of
these notices in the summer of 1855, and described in the
autumn of that year at the Glasgow meeting of the British
Association. Both tracts occur on that side of the sandstone
district which is nearest to the " granite nucleus ;" and the
occurrence of the rock here is thus intimately related to the
outburst of the central granite, in its abnormal position —
close to the outer border of the upper slate, and to the base
of the sandstone formations — already pointed out as the
great leading peculiarity of Arran. The relation of these
granites to that of the nucleus, and to the adjoining strata,
is shewn in the annexed ideal section from N. to S. across
the granite nucleus (fig. 6) : —

THE PLOVERFIELD GRANITE.

17

TJie Ploverfield Granite.

11. The Ploverfield granite
occurs amid strata which un-
doubtedly belong to the carbon-
iferous' system. The tract is
situated on the west side of
Glen Cloy, and that branch of
it called Glen Dhu. The hills
here are the highest portion of a
long ridge running up south-
west from the plain of Brodick,
and dividing Glen Cloy from
Glen Shirag. At the origin of
this ridge are beds of carbon-
iferous limestone with fossils,
under the north wall of the
enclosure of Brodick Church.
On the upper slope of the ridge
beds of limestone and shale
also occur in the sandstone, not
far from the north base of the
Windmill Hill — a high, elon-
gated, narrow ridge, steeply
overhanging Glen Dhu, and
composed of quartziferous por-
phyry. On its south-west base
the granite first appears, sepa-
rated, in some places at least,
from the porphyry by a narrow
band of altered sandstone.
Thence, keeping at a high
level, it extends along the
hill slopes at the head of Glen
Dhu, and terminates southwards
against a ledge or low cliff'

c

-S ro 2
PH ra 'S

10 «
•3

.£F 5j

o

„ is

s »

a y> aT

2 «--g

be .^ 5

III

•» °° «•

« - S

5 "- 5

»- o 'C

«a & 3

18 GEOLOGY OF ARRAN.

of syenitic greenstone, a little west of the top of the
Blackhill — a high, bold point separating Glen Dhu from the
upper part of Glen Cloy. The top and front of this hill toward
both glens are composed of altered sandstone, resembling
quartz-rock, the change of structure being, no doubt, due to
the proximity of this syenitic greenstone. The ledge or cliff
teems to be the northern front of a large dike ranging be-
tween west and north, and cutting off the granite. To the
south of it the hills are composed of sandstone, with occa-
sional knolls of overlying common trap. Westwards, the
granite extends across the moorland tracts, which lie between
Glen Cloy and a high ridge forming the watershed or axis of
the island at this part. It is seen in a few rocky points, but
chiefly rising through the sandy substratum of the peat,
where this has been worn away, or in the beds of the small
streams. As we advance westwards, it becomes gradually
intermingled with syenite and porphyry: but the relation of
the three rocks cannot be made out, owing to the nature of
the ground. They appear to alternate, the granite diminish-
ing in quantity toward the west. On the summit of the
highest ridge or watershed loose pieces of granite strew the
surface, mixed with syenite and porphyry. Bosses of the
latter rise through the broken-up masses of the sandstone,
forming the main body of this ridge. The surface is also
strewn with masses of pitchstone, and of the coarse-grained
granite of the northern mountains. We must thus be con-
tent with a hypothetical boundary for the Ploverfield granite
on the west. Its southern boundary does not reach so far
as a line joining the heads of Clachan glen and Glen Cloy,
sandstone being seen continuously between them. On the
north side its boundary runs from the west base of the Wind-
mill Hill to a point considerably to the west of the summit
level of the String road. Here a high ridge on the south side
of the glen, and west of the watershed, presents granite in its
upper part; the lower portion is composed of a dark-red and
hard sandstone; the rocks are seen in close proximity, the

THE CRAIG-DHU GRANITE. 19

sandstone ranging along the lower part of the hill, the granite
on the upper part and southwards over the summit. The con-
tact is seen at only one point, which is on the hill side due south
of the summit level of the road. Here the granite seems to rise
through and overlie the sandstone; the rocks are intimately
blended and assimilated in structure; the sandstone is less dark
and more crystalline, while the granite has more the look of a
crystalline sandstone than is usual with it. The limits above
traced mark out an area much lai-ger than that hitherto
assigned to the granite. The weathered surface of the rock
over all the area is generally white; but the prevailing
colour in a fresh fracture is red. It is fine-grained, with
little mica in proportion to the quartz and felspar, and is
very similar to the fine-grained variety of the interior.
Drusy cavities often occur, lined with crystals of quartz and
felspar, and large crystals of mica are not infrequent. Mr.
Eamsay states (Guide to Arran, p. 1 2) that " It sends forth
veins into the adjacent sandstone; while specimens of the
sandstone, much altered by the effects of intense heat, may
even be found enclosed in the granite." We have not
noticed these cases of intrusion, nor are the precise spots
mentioned. We have, however, seen such in connection
with the porphyry of Windmill Hill, Jfco which the granite
succeeds on the west, and also in connection with the new
granite tract to be next noticed.

T/te Craig -Dhu Granite.

12. The other granite tract lies on the south side of the
Shiskin road, south-east of the farm-house of Glaister.
Here the hill, whose base is skirted by the road all the way
down from the " String," overhangs the valley of Mauchrie
water in a steep cliff called Craigmore, Craig-Dhu, or The
Corby's Rock. This cliff is the outer edge of a small plateau
or table-land, cut off from the higher ground behind, towards
Ard-Bheinn, by a deep hollow, which completely isolates it.

20 GEOLOGY OF ARRAN.

The summit is 500 to 600 feet above the valley, and is more
than a quarter of a mile long, and from 250 to 300 yards
broad. It descends steeply towards Shiskin on the south-
west, and slopes gradually north-east towards Glen Loig.
The summit and sides of this plateau are formed of fine-
grained granite, very similar to that of Ploverfield. The
base of the cliff towards Mauchrie Water is covered by a long
talus of granite blocks and smaller fragments, reaching to
within 50 or 100 yards of the road, and appearing even at
that distance of so different an aspect from fallen masses of
sandstone that it is surprising it was not noticed till our first
visit to that side of the island in 1855.

The granite here seems to rise either through the old red
sandstone or at the junction of this rock with the carbonife-
rous strata. The granite is nowhere seen in situ at a low
level ; the talus before mentioned obscures the rocks along
the base of the hill; and the ground by the roadside arid
along the valley is deeply covered with alluvium. At some
spots, however, rocks are seen apparently in situ. They are
not of a very marked character, but seem to be chiefly sand-
stone of the old red formation, greenstones and trap porphyries,
the sandstone having assumed a subcrystalline metamorphic
aspect from the intermixture. But at a high level on the
west, south, and east sides of the plateau, the granite is seen
to rise through a coarse conglomerate ; and numerous con-
tacts are observable. These are highly interesting, and
clearly indicate the intrusion of the granite subsequently to
the formation of the conglomerate. The base of this con-
glomerate is a coarse sand, and the imbedded fragments
sandstone, quartz, and granite. The base is highly indurated,
and assumes a porphyritic structure ; the sandstone is
rendered crystalline, and the quartz has been fused, and
reconsolidated into a substance resembling porcellanite. The
fragments of granite are of an elliptic form, less rounded
than the quartz, and are exactly like the adjoining mass of
granite in structure and component parts. Whence have

RELATIVE AGE OP THE GRANITES. 21

these granite fragments been derived? From the body of
fine granite among the northern mountains, or from the
adjoining mass itself? Mineral structure does not enable us
to determine — the two rocks are so similar. If from the
former source, then we must conclude that the granite of the
interior was elevated so as to be exposed to disintegrating
causes, while the conglomerate was forming; in which case
granite fragments ought to occur abundantly in the sand-
stone conglomerates ; but this is not found anywhere in
Arran — a fact noticed by all observers. Even, here the
fragments occur only in close proximity to the granite itself.
Must we not then rather suppose that pieces of the granite
adjoining, when this rock was erupted in a fluid or semi-fluid
state, were injected among the outer strata of the conglo-
merate, also fused by the contact, and so became imbedded
in these strata only ?

Sir Charles Lyell, in discussing the relative age of the
Arran granites (El, p. 720, edition 1865), favours the view
stated in the last sentence. Having quoted the above passage
from our former edition (3rd, 1864), he observes, — -"I have
seen a similar junction in Caithness, of which Sir R.
Mxirchison, in 1827, and again in 1828 jointly with Pro-
fessor Sedgwick, has given a faithful description. Close to
the point of contact of certain oolitic sandstones, shales, and
limestones, in the Caithness cliffs, a breccia occurs containing
granite fragments, mixed with those of the invaded rock.
The granite, they say, appears as if it had been mechanically
driven in among the shattered and altered strata."

This case, unknown to me till I saw it quoted as above,
strongly supports the view which I have taken, that the
imbedded fragments at Craig-Dhu form no exception to the
general rule regarding the Arran conglomerates.

Relative Age of the Granites.

13. Granite, then, occurs in Arran, in three disconnected
tracts, — amid slates in the northern mountains, in the old red

22 GEOLOGY OF AKRAN.

sandstone of Craig-Dhu, or at the junction of this rock with
the carboniferous strata, and also amidst these carboniferous
strata at Ploverfield. In each of these positions it must be
regarded as posterior to the rock which encircles it, since it
is intruded among these sedimentary deposits and produces
a marked alteration upon them along the planes of contact.
We have, therefore, now to consider the question of age.
Are the three granites of three distinct ages coi-responding
with those of the strata among which they intrude ? or were
they erupted simultaneously, so as to pierce through the
three formations during one and the same period of disturb-
ance 1 In other words — and this view narrows the question
— since the Ploverfield granite is clearly of later origin than
the sandstones of Windmill Hill, and the shell limestones
subordinate to them, were the granites of the nucleus and of
Craig-Dhu erupted at the same time with it ? or does their
injection among the strata, and elevation to the day, date
back to an earlier period ? The close proximity of the Craig-
Dhu granite to the border of the carboniferous formations,
if it be not actually enclosed in these, evidently points to an
identity of age with that of Ploverfield, and renders their
simultaneous eruption extremely probable. How then is
the granite of the nucleus related to the old red and
carboniferous formations? It has been long established, and
is well known that it everywhere throws powerful veins into
the encircling slate-band, greatly altering this rock along the
line of junction, producing banding and contortion, and dis-
turbing its stratification. In some vertical sections, as at
Torneadaneoin, and the hill-sides north of Glen Catacol,
we have slate above and granite below, with numerous alterna-
tions where the two rocks approach. Hence we infer that
the granite was injected in a melted state amid the already
formed strata of slate. But, further, it was suggested long
ago by Murchison and Sedgwick, in their celebrated paper
on Arran (Geol. Trans., vol. iii., second series, 1835), that
the bed of limestone on the north front of Maoldon may have

RELATIVE AGE OF THE GRANITES. 23

once been continuous with the Corrie stratum, and that its
actual position is due to an upthrow by a protruding mass of
granite advancing from the central mass east of Goatfell. If
this be admitted, it would make the intrusion of the granite
posterior to the deposition of the carboniferous formations,
and so render probable an identity of age between the
Ploverfield granite and that of Goatfell and the nucleus.
We think, however, that this view is liable to question; we
often find exactly similar bands of limestone amid the strata
of coal sandstone, without any evidence of former continuity,
or such cause of disturbance; and the amount of vertical
displacement implied in the supposition could hardly have
taken place here without a fracture of the crust, and the
appearance of granite on the surface, — so narrow is the band
of sedimentary strata superimposed upon the granite. A
positive conclusion, then, seems scarcely justifiable from this
case. A stronger presumption is derived from the high
angle generally assumed by the sandstones, where they
approach near the gi-anite, on account of the narrowness of
the slate band ; and from the degree of metamorphism which
in such situations they exhibit, a good example of which
occurs at the junction in the burn of the White Water, above
Corrie, where a gradual passage takes place from slate to
sandstone, clearly the effect of metamorphism, by the heat to
which both were subjected. The facts all tend to shew that
the granite was injected and elevated after the deposit of the
old conglomerate, and that the entire slate stratum on the
east or Corrie side was in a plastic state, under the influence
of the heat which fused the granite.

Viewing all these facts in connection with the general con-
formability of the carboniferous strata to the old red sand-
stone, and the gradual transition from the one series to the
other, observed in several places, it is even probable that the
injection of the granite took place after the deposit of the
carboniferous formations; and that therefore the granites of
the three disconnected tracts may be all of one age, or belong

24

GEOLOGY OF ARRAN.

to the same period of disturbance. But as the granite of the
nucleus is nowhere seen to alter the carboniferous formations,
while it certainly does, as above stated, alter the old red
sandstone, it is quite possible that these carboniferous strata
may have been deposited upon the old red sandstone during a
period subsequent to the irruption of the granite. But this
irruption took place in hypogene depths, not only prior to
the elevation of the island above the waters of the primeval
ocean, but while the granite was yet enveloped by the
mantling slate rocks, and perhaps also by the later forma-
tions, as shewn in the annexed cut (fig. 7), in which a is the
granite yet enveloped by the slate; b, slate; and c, conglo-
merate, with slate fragments, but none of granite. It is

a Fig. 7. a-

a a, Granite; 6, slate; c, old red sandstone.

obvious, as already pointed out in Art. 9, that these secondary
strata have not derived the detrital materials of which they
are made up from the disintegration of the granite ; this rock
•was yet protected from disintegration by its mantle of slate ;
and the Old Red derived its materials from other, and some
of them remote sources. An extensive disintegration and
denudation may even have gone on for a long period, ere yet
the strata were injected by the melted granite; for in several
places the conglomerate, partly made up of slate fragments,
is altered as well as the underlying slate ; and besides,
the extreme narrowness of this band on the east renders' it
very improbable that portions of the injected veins, with
adhering slate, should not be found in the conglomerate, if
the injection had been prior to the denudation and to the

RELATIVE AGE OF THE GRANITES. 25

deposit of this latter rock. Such a negative argument, how-
ever, has not much value.

The elevation of the central granite mountains to their
present height may have been a gradual process, during the
continuance of which, in waters constantly becoming
shallower, the strata of slate may have been exposed to
further extensive denudation, which, joined to various at-
mospheric influences, afterwards acting, would give their
present form and outline to the jagged ridges of the northern
mountains. Long before this elevation took place, the
granite, under the pressure of the superincumbent slate, and
perhaps of the newer formations also, had acquired its
crystalline structure by the slow passage of its heat of fusion
into the adjoining strata; and most probably it was quite
solidified anteriorly to its elevation, so that it was protruded
in a solid form.

14. The agent in this protrusion may have been a newer
granite, produced beneath an older. Let fresh accessions of
molten matter — the matter of granite — be slowly and con-
stantly transfused from the nether depths, amid the basement
portion of the older granite, already cooled and crystalline
above, while fused below by contact with the molten mass —
this latter will expand, and perhaps laterally extend the
former, and raise it in a solid form. Thus a great upward
movement might be produced, forming the high mountain
nucleus of the north, and at the same time elevating and
contorting the strata of slate and sandstone resting on the
flanks of the older granite, and in some places perhaps even
inverting the dip of the slate, as being subject to a greater
strain, and more likely to yield en masse, without disruption ;
while the sandstones of the southern plateau, remote from
the focus of intensity in the upheaving force, would be
elevated from below in more horizontal strata. The newer
granite from below might likewise impenetrate the older,
overflow it in places, and spread out over it; or even break
through and impenetrate other strata beyond the limits of

2G GEOLOGY OF ARRAN.

the older granite. Formed under such conditions, this later
granite might be expected to differ in structure, if not also
in composition, from the older granite or other rocks invaded
and displaced by it. Such differences we know do actually
exist between granites in the Alps, Andes, and other
localities, which can be clearly proved to be of different
ages. Now, we have two such granites within the area of
the mountain nucleus, the coarse or under-lying, and the
fine-grained or over-lying (Art. 6). "With this latter, Mr.
Ramsay's Ploverfield and our Craig-Dhu granite exactly
agree (Art. 11, 12). Is it not, then, probable that these
three, the fine-grained of the nucleus, and those of Plover-
field and Craig-Dhu, belong to one period of disturbance —
that they were simultaneously injected amid the rocks which
now enclose them, at a period subsequent to the deposit of
the carboniferous strata? This conclusion is rendered highly
probable by the character of the contact between the fine
granite and the slate, on the south boundary near Loch
Ghnuis, discovered by us in 1864, and to which reference
has been made (Art. 6). The two rocks adhere firmly, the
slate is altered in the usual way, and invaded by ramifying
veins, in the same manner as the coarse variety invades
the slate in other places. At several yards back from the
junction a few masses or large blocks of the coarse-grained
variety are seen, and in two or three cases were observed to
be penetrated by veins of the fine kind ; in othei-s the two
kinds were irregularly associated in one block, in a way
suggesting either injection of one into the other, or a simul-
taneous crystallization under varying conditions, the true
relations being masked by the extensive decomposition which
lias affected both. The junction here was traced for upwards
of a mile, and the coarse and fine kind found to be irregularly
associated throughout, the fine predominating, and the coarse
often occurring in such positions that it was difficult to
determine whether it was in situ or transported. The
contortions of the slate are very striking along the boundary.

RELATIVE AGE OF THE GRANITES. 27

The appearances here, in connection with the other relations
already noticed, certainly favour the idea of a late intrusion
of the fine granite. Some importance, however, must be
attached to the occurrence of the rounded pieces of fine-
grained granite enclosed in the conglomerate, close to the
Craig-Dhu granite (Art. 12). If our explanation of their
occurrence there by injection be deemed unsatisfactory, then
will not the above conclusion hold; and it must be admitted
that, before the irruption of the Craig-Dhu granite, the fine-
grained variety of the interior must have been elevated,
stript of its slate covering, and exposed to degradation.

15. What then, it may be asked, is the conclusion which
we favour, and to be finally drawn from these various and
somewhat conflicting statements? The discussion may seem
tedious and unimportant to many; yet we hope it will not
be without its use to the student and future inquirer; and
as several of the facts are new, it may have some value in
the eyes of the many geologists in this and other countries,
who have either written upon the subject or take a lively
interest in the physical history of this extraordinary island.
The question of relative age is, we hope, at least much
narrowed by these statements, if it cannot be held as quite
settled.

The various possible conclusions may be set forth, by way
of recapitulation, as follows : —

(1). The oldest rock in the island is the slate; it was
elevated and exposed to disintegration ere yet the granite
had reached the surface, or probably been even intruded
among the slate strata. These had, however, been invaded
by trap dikes ; for we find at the mill-dam, on the ascent of
Goatfell, a basaltic dike intersecting the slate, and itself
intersected by a granite vein !

(2.) The old red sandstone, and the carboniferous sandstones
with their intercalated limestones and coal strata, were
formed before the granite was exposed to disintegration, the
only fragments of this rock yet found being those in close

GEOLOGY OF ARRAN.

proximity to the Craig-Dhu mass, of which they are probably
injected pieces, and not derived from the disintegration of a
granite already exposed.

(3.) The injection of the granite of the nucleus, whether
coarse or fine, in a melted state, amid the slate strata, was
certainly posterior to the deposit of the old red sandstone,
and may have been posterior also to that of the carboniferous
strata.

(4.) If the granite of the nucleus be thus of later age than
the carboniferous strata, then may all the granites be of one
age, if such differences of mineral structure, or aggregation
of parts, can be admitted to exist in contemporaneous
granites.

(5). But as the coarse-grained granite cannot with certainty
be pronounced newer than the carboniferous strata, while
the Craig-Dhu and Ploverfield granites undoubtedly are so,
then we may have two ages for these outbursts — one for
that of the coarse granite, and another for that of these
latter.

(6.) The constant character of the fine-grained granite of the
interior, through a considerable area, its impenetratiou of the
coarse, and of the slate in veins, point to a later origin than that
of the latter; while its almost perfect identity in structure and
arrangement of parts with the other two granites render
very probable the contemporaneity of these three, and their
posteriority to the coarse-grained variety. This conclusion
seems now warranted by the facts.*

T/te Trappean Rocks.

16. Arran is extremely rich in rocks of this class; most of
the known species occur, and also those numerous varieties
by which these graduate into one another. They form great

* The conclnsion stated in the text has been embraced by Sir Charles
Lyell (loc. cit. ) on a review of the evidence set forth in our last edition,
as well as from his own knowledge of Arran.

THE TRAPPEAN ROCKS.

29

overlying masses, capping the sandstone of the southern
plateau, and rising into the highest hills of this division of
the island. They are interposed amid the sedimentary
deposits, in huge sheets or beds conformable to the stratifi-
cation, and cut through all the rocks alike, from the lowest
to the highest, in vertical or slightly inclined dikes, which
range continuously across great horizontal distances. These
dikes are never observed to wedge out downwards; and no
doubt they descend to the level of cooled down lakes of
once molten matter, concentric with the crust — the common
' source whence they all proceeded, and whose vents or outlets
these dikes once formed, in past stages of the earth's history,
when the various rocky materials were elaborating. The
pressure of an ocean of great depth, or that of other strata,
amid which they were poured out, gave these various igneous

Fig. 8.
Relation of the Igneous and Sedimentary Rocks.

products that density and compactness, which constitute
almost the sole differences between them and the modern
products of fire thrown out under the pressure of the atmos-
phere only. Their differences among themselves are due, less
to a difference in mineral or chemical composition than to
the arrangement of component parts which would be given
by different rates of cooling from a state of fusion.

The trap rocks of Arran may be arranged in three classes,
according to their composition, —

The Felspathic, comprising porphyry, claystone, compact
felspar, and pitchstone.

The Hornblendic, as diorite and amygdaloid.

The Augitic, as basalt and dolerite.

oO GEOLOGY OF ARRAN.

These are all intimately connected, one species often
passing into another by regular gradations, and they are all
found in the same relative positions with respect to the
sedimentary strata. Among themselves they do not preserve
any order of succession, nor do they occupy separate areas,
so that their continuity cannot be reckoned on through a
considerable space. Nor can they be indicated on a map by
distinct coloui's, so intimately are they blended with one
another. Basalt, diorite, amygdaloid, and porphyry are by far
the most abundant, as well in overlying masses as in dikes
and interposed beds. Porphyry is next in abundance, and
occurs in all these positions; pitchstone alone has not over-
flowed the surface, and occurs only in dikes and beds. The
overlying masses are limited to the southern section of the
island, while dikes occur everywhere, not, however, with the
same frequency in all parts. On this subject Professor
Phillips was, we believe, the first to offer a good generaliza-
tion— " Dikes are most abundant at some distance from the
granitic centre. At Corriegills, at Lamlash and Tormore,
they are exceedingly abundant in the red sandstone, while
in the north-eastern face of the island, where that rock is
nearer to the granite, fewer dikes appear; and about Loch
Ranza the slate is still less divided by them. Perhaps we
may venture to add another generalization, — viz., that these
dikes are most abundant beyond the line of violent flexure
of the strata from their horizontal position. After measuring
with care the directions and breadths and noting the
characters of forty -four dikes, chiefly of gieenstoue, between
Brodick and Lamlash, and also those at Tormore, it did not
appear to us that any other dependence of the direction of
these dikes upon the local centre of the granitic eruption
could be traced." — (Man. o/Geol., 1855, p. 505.)

Rocks of the felspathic type, which are most closely allied
to granite, seem to have no more intimate relation to the
granitic centre than have those of the hornblendic or augitic.
The largest body of porphyry on the island is that on the

THE TEAPPEAN ROCKS. 31

south-west coast; the next in extent is that of Dunfion, over
the Corriegills shore. A similar rock occupies a small space
on the Windmill Hill, over Glencloy, in close connection with
the Ploverfield granite. A different variety forms the bold
pi-ecipices of Drumadoon, on the west, and the principal mass
of Bennan Head, on the south. The largest body of claystone
forms the middle and upper portions of Holy Isle, and has a
thickness of nearly 900 feet ; extensive beds and dikes of the
same substance are met with in Lamlash river and at Black-
water-foot; lesser veins and beds in many other places. All
of these cut through or overlie the carboniferous formations of
the southern section of the island; the northern section, the
i-egion of granite, slate, and old red sandstone, is almost
devoid of these felspathic rocks ; a few dikes only are met with ;
almost all those in this tract being of hornblendic or augitic
rocks. The pitchstones also, exclusive of those in the gi'anite,
are almost all met with in the neighbourhood of Brodick,
and towards Mauchrie Water on the opposite shore. These
various felspathic rocks thus seem to correspond pretty nearly
on opposite sides of the island, and to have no relation in
their position to the granitic nucleus. Neither does there
seem to us any good foundation for a generalization put for-
ward by some writers on Arran, that rocks of this type are
more abundant on the western than on the eastern side of
the island.

Overlying rocks, hornblendic or augitic, occur chiefly over
the central and south-eastern portions of the southern
plateau, south of the parallel of Lamlash. North of this line
they have little horizontal extension ; they merely cap the
sandstone in isolated knolls or narrow bands of inconsiderable
thickness. The details regarding these, as well as the fel-
spathic rocks, will be seen upon the map, and will be more
fully noticed in the several " Excursions " which follow.
Under the same head we shall notice the changes made by
the dikes on the adjoining rocks.

In Arran, as elsewhere, almost all the dikes are simple —

32 GEOLOGY OF ARRAX.

that is, composed of one kind of rock; a few only are com-
posed of parallel bands of different substances. By far the
greater number, in every part, consist of common trap — that
is, some variety of greenstone or basalt.

17. M. Necker, who visited Arran in 1839, carefully
measured the direction by the compass of a great number of
dikes seen along the eastern shore and in the interior, and
laying these directions down upon a map, he identified a
great many of them at remote points. Such continuity has
been often made out through distances greater than the whole
length of Arran ; but we know of few tracts pervaded by
dikes, where the same caution is required in drawing such
conclusions, on account of the great number of dikes, their
varying directions, and the undulations to which the same
dike is subject. The prevailing direction is towards the
north-west and north-east quarters, and nearly within the
limits between which the magnetic needle is known to vary.
Yet are there many which range without these limits, and
not a few have a direction nearly due east and west. Most
of the dikes are vertical; those which are inclined to the
horizon seldom pass an angle of inclination of 20°. M.
Necker estimates the number of dikes between Loch Ranza
and King's-cross point at 200; the number to the south of
this,' on the east coast, at 144, making a total of 344. The
remaining portion of this estimate we give in his own
words : —

" Mais 1'evaluation precedente ne comprend que les dykes
de la surface d'une inoitiS environ de 1'ile; tout 1'int^rieur de
la partie meridionale n' y est pas compris, non plus que la
cote N.O., ni le groupe granitique de Ben Vearan entre cette
c&te et la rividre Irea; et quoiqu' il soit connu que 1'interieur
des terres renferme toujours moins de dykes que les cdtes, et
que la cdte N.O. est en g6n6ral tres d^pourvue de dykes,
quoiqu' enfin cette moitie de 1'Ile soit bien plus petite que
cette que j'ai parcourue; omettant ces circonstances je por-
terai pour elle un nombre £gal a la premiere, soit 344, form-

THE TRAPPEAN ROCKS. 33

ant un total de 688, ou, en nombre rond, de 700 dykes de
trap dans la totalite" de 1'ile de Arran. Doublant meme
encore ce nombre si Ton vouloit, pour y comprendre tous les
dykes caches par les bruyeres vastes et etendues dans 1'in-
terieur, par les greves de sable sur les rivages, ou places dans
des recoins inaccessibles des ruontagnes, on n'arriveroit pas
encore au nombre de 1500; et pourtant en parlant de telle
ou telle cote, de telle localite d' Arran, il est souvent echappe
a ceux des geologues qui ont decrit Arran, a moi-meme peut-
etre tout le premier, de dire qu' on y voyoit des innombrables
dykes de trap. Or, je crois avoir maintenant montre que
loin de ne pouvoir etre compte"s, on peut a present concevoir
1'esperance de voir chacun des dykes de cette lie individuelle-
ment etudie, nume"rote, decrit et enregistre dans un catalogue
descriptif et raisonne, analogue a celui que j'ai aujourd' hui
1'honneur de mettre sous les yeux de la Societe Royale." *

Mr. James Napier, of Glasgow, has published a short
paper on the dikes between the bays of Broclick and Lamlash
in the Edinburgh New Philosophical Journal, New Series,
vol. ii. No. 1, July, 1855, accompanied by a map, on which
the dikes are laid down. He reckons altogether fifty-four
dikes as visible along the shore, but considers that not a few
may have escaped his notice. " Struck," says Mr. Napier,
"by the large number of trap dikes cutting through the
sandstone, in a direction at right angles to the sea-line, it
occurred to me that if such dikes continued round the coast
to Lamlash, and still at right angles to the sea-line, they
must in all probability have proceeded from a common
centre, lying somewhere between the two bays." To test
this idea by observation, he measured and marked down the
position of every dike, and the result confirmed his " antici-
pation, that they proceeded from one, or possibly from two

* Documents sur les Dykes de Trap d'une Partie de File d' Arran.
Transactions of the Royal Society, Edinburgh, vol. xiv. Part 2, 1840,
p. 684. The paper is a model of patient and generally accurate re-
search.

D

34 GEOLOGY OF ARRAN.

centres." A similar idea would be very likely to occur on
examination of M. Necker's map, which certainly Mr. Napier
had not seen, else he would have mentioned it. The notion
of radiation from a common centre we do not, however, find
alluded to in M. Necker's paper. Mr. Napier seems dis-
posed to assign two centres — one for the felspathic dikes, and
another for the hornblendic — both lying inland towards the
Lamlash road. Prolonging the directions of the two prin-
cipal felspathic dikes on the shore, he finds that they would
meet near the claystone quarry on the Lamlash road, about
a mile from Springbank ; and here he would place the fel-
spathic centre; the hornblendic he does not so definitely fix.

Now, analogies in support of this view can certainly be
drawn from districts of recent volcanic action, where fissures
radiating from a vent, or focus of disturbance, are seen to be
filled with basaltic lava and other igneous products; and the
same may doubtless have occurred in the case of the plutonic
rocks ; but the evidence for it in Arran we cannot consider
sufficient There are many exceptions to the rectilineal*
course of dikes here as in other places; some of the dikes
converge towards the Corriegills shore, and the largest runs
a long way parallel to it, while one at least re-appears far
inland beyond the place of the supposed focus. Besides, so
far from " the whole of the hills between Brodick and Lam-
lash being composed of trap," this rock is, in point of fact,
confined to a narrow and thin capping along the highest
ridge between the two bays; and Mr. Napier has overlooked
the great outburst of porphyry at Dunfion, which has a
manifest relation to the felspathic dikes on the shore, as well
as the numerous masses of claystone intercalated amongst
the sandstone strata along the northern slopes west of
Corriegills.

On these grounds we cannot admit that this speculation
has much value ; the apparent radiation arises from the cir-
cumstance already mentioned, that the vast majority of the
dikes range between the points of extreme magnetic declina-

GLACIAL ACTION. 35

tiou east and west. Some other suggestions, however, of this
paper have great value — those, namely, of a chemical nature,
referring to the different degrees of fusibility and rates of
cooling among trap rocks, which, if experimentally estab-
lished, would elucidate many points still obscure in the
natural history of the ancient products of plutonic fires In
a later paper, describing the Whiting Bay dikes, Mr. Napier
forms a much higher estimate than M. Necker of the total
number on the east coast (Trans. Phil. Soc. Glas., vol. iv.
p. 321).

Glacial Action.

18. Till the publication of the present work, glacial action
had not been recognized in Arran ; appearances had, indeed,
been observed by Dr. MacCulloch, in regard to scattered
blocks and " alluvial mounds," of which the ice theory, not
thought of in his day, now offers the best explanation ; but
the island had not been examined with the view to discover
direct evidences of the action of ice, as in the striation,
polishing, and rounding of rocks, the transport of boulders,
and the composition and other characters of the superficial
accumulations. Some general results of our inquiry will
now be stated; details being reserved for the notices regard-
ing particular tracts, which will be given farther on.

The remarkable peak of Cior-Mhor has been already
pointed out as the geographical centre of the northern group
of mountains (Art. 2). From its base the four great valleys
of the island — Sannox, Rosa, lorsa, and Eas-an-Bhiorach —
radiate in all directions, their extremities opening on the
seaward belt of low land. If glaciers ever existed in Arran,
under the subarctic climate to which Scotland was once sub-
jected, these central heights must have been the seat of the
snow-fields which fed them, and the radiating valleys the
channels down which the viscous mass of glacier ice must
have pushed forward to debouch upon the low ground, and
melt under a higher temperature. On the sides, then, and

36 GEOLOGY OP ARRAN.

towards the openings of these valleys, we should expect the
effects of glacial action to be most distinctly traceable in the
striation and polishing of the subjacent rocks and transported
masses, and in the formation of lateral and terminal moraines.
Many broad surfaces of the natural rock are exposed both
on the sides and in the bottoms of these valleys, favourably
placed for receiving such impressions under the grinding
action of a descending mass ; yet the cases of striation and
polishing are not numerous. A granite surface is, how-
ever, very unfavourable for the preservation of such mark-
ings, especially the Arran granite, which is generally of such
structure as to be subject to rapid disintegration. The slate
is better fitted to retain impressions of this kind, its tough-
ness and fine-grained structure rendering it less liable to
decomposition; but it is seldom exposed in favourable situa-
tions, and is rarely found striated. Granite bosses in the
glens, and on many of the lower ridges, have that peculiar
rounded character, due to the action of ice, to which the
term "roches moutonn^es" has been applied; but perhaps
few of the cases ought to be decidedly referred to glacial
action, on account of the peculiar spherical structure so often
assumed here by granite on the large scale. On the slate
ridges, however, beyond the granite border, on the old red
sandstone, and on the trap of the southern plateau some well-
marked cases do occur. Other evidences, scarcely less satis-
factory, are of frequent occurrence. These are the terraces
and mounds of transported materials on the sides and at the
openings of the glens, the boulder clay deposits, and the
dispersed blocks in every part of the island.

The terraces and mounds consist of earth and rounded
masses of rock of each particular glen, irregularly mixed, with-
out reference to weight, and in such situations that they
could not have been brought together by existing river action,
being much above the level of the streams which now tra-
verse the valleys. They are most probably referable to
glacier moraines of two classes — the lateral and terminal —

GLACIAL ACTION.

37

formed by masses of rock descending from the highest peaks,
and thrown to either side by the movement of the ice, or
deposited at the extremity of the glacier when the ice melted.
The former have been much modified by torrents entering
from the sides, after the glaciers disappeared, and now pre-

Fig. 9.
Lateral Moraines.

sent but detached mounds. In some glens, however, as
Sannox, terraces yet remain complete, but not of great extent.
The terminal moraines are better marked, fine examples being
visible at the mouths of Glen lorsa, Glen Catacol, Glen
Rosa, and others. The remarkable terraces at the opening

Fig. 10.

Terminal Moraine.

of Glen Catacol skirt the valley on the south-west at a height
much exceeding any level the stream could now reach by the
joint effects of floods and high tides, and indeed surpassing
that which it could ever haver attained, even when the sea
covered the present maritime belt or terrace. We are,
therefore, inclined to regard these mounds and terraces as
terminal moraines, modified in their outlines by floods^ tides,
and ordinary river action.

Still more remarkable are the lofty terraces at the mouth

GEOLOGY OF ARRAK.

of the river lorsa ; they are far more striking, indeed, than
anything of the kind in Arran. They consist throughout of
transported materials, some of the rocky masses being very
large; the sides are steep and the summits usually flat; and
the height of the highest is sixty or seventy feet above the
river, and at least thirty above the ancient sea level just

Fig. ll.

alluded to. Speaking of these (lorsa, and other such mounds)
in reference to river action, MacCulloeh remarks : — " The
origin of such alluvia is very obscure — a few may have been
deposited in particular situations by the same waters which
are now removing what they formerly laid down ; while in
other cases it is impossible to assign any mode of action by
which this double and opposite effect could have taken place
from one agent. . . . The quantity and quality of the
materials, their extremely rounded forms, the nature and
permanence of the hills above, and the want of a regular
gradation of size in the stones from the bottom upwards,
seem to show that other causes [than river action] of a tran-
sient, and probably of a diluvian nature, have in distant times
generated these deposits, which have been subsequently acted
on by the stream concentrated on the bottom of the glen by
the form of the ground" ( Western Isles, ii., p. 335; 1819). The
difficulties of these cases had thus presented themselves to
the mind of this distinguished geologist, and he offers the
best explanation that could then have been given.

Similar mounds occur, but not in the terraced form, at the
openings of Glen Sannox and Glen Rosa, much elevated

GLACIAL ACTION. 39

above the river beds. Some way up the latter, also, there
are remarkable mounds, in a situation where, from a great
bend in the glen, we should expect a moraine to be thrown
down. Many geologists may, however, favour the idea that
the mounds and terraces in question were formed when cur-
rents swept these glens, during the gradual elevation of the
land. That such elevation may have been a long-continued
process we have already seen reason to suppose (Art. 13, 14);
and the effect must have been a general disturbance in the
sea-bed, which, joined to the action of tides produced then
as now, could not fail to give rise to currents of considerable
force. Where these met the sea, towards the mouths of the
glens, banks and terraces may have been thrown up; or a
sudden elevation of the land of a cataclysmal character may
have given origin to long-continued currents of sufficient force
to transport large blocks, and to throw down a promiscuous
deposit, such as we find in the mounds and terraces of the
mountain glens. Such sudden elevations of a range or group
of mountains are still regarded by many geologists as the true
explanation of the " diluvial phenomena ; " and it was such
probably that Dr. MacCulloch had in view when he spoke, in
the passage above quoted, " of other causes of a transient and
probably of a diluvian nature," as giving origin to the remark-
able accumulations at the mouths of the Catacol and lorsa.

19. The dispersed blocks present phenomena still more
curious and of much more difficult explanation. They are
almost exclusively granite, a very few only of slate being
found. They are scattered in great numbers over every part
of the island, and are often of enormous magnitude. They
are most abundant and largest in the vicinity of the granite
nucleus, as about Corrie and the shore at Corriegills; and
generally less numei-ous, and of smaller bulk, in the remote
southern districts. Occasionally, however, some very large
ones occur even there. They are limited to no particular
locality, but occur alike in the valleys, on the summits and
northern and southern slopes of the hills, in situations to

40 GEOLOGY OF ARRAN.

which they must have passed across deep and narrow glens.
They are found also isolated on the Holy Isle, which is
separated from the mainland by a wide bay and two deep,
navigable channels. Blocks of the coarse-grained variety are
much more numerous than those of the fine. The latter,
indeed, are in a great measure limited to the tracts on which
Glen lorsa, the principal seat of this variety, opens towards
the south ; and this fact, in connection with the more sparing
distribution of the blocks along the northern coast, on which
but one glen with a narrow opening debouches, than in other
parts of the island, shews that, though the dispersion has
been quaquaversal, it has been to a large extent determined
by the direction of the valleys. Dr. MacCulloch, who has
noted the leading facts regarding the dispersion of the granite
blocks with great accuracy, though imperfect in many of his
details, closes his account with the following observations : —
" None of the blocks have the marks of a distant origin ; all
have the characters of the granites of the adjoining moun-
tains, characters sufficiently distinct from those of almost all
the granites of Scotland. . . . No situation, perhaps, has
been pointed out where the origin of the travelled blocks is
more obvious, or their new position more difficult to compre-
hend, without assuming considerable revolutions of the sur-
face of the land over which they have passed. . . . The
compact and solitary position of the fixed mass of granite,
the identity of the materials of this mass with that of the
travelled stones, the gradual diminution of these as they
recede from the parent rock, and the insulated position of
the whole, render their origin indubitable, and present to the
geologist a spot, on the changes of which he may speculate,
with the certainty that he has before him a set of incontro-
vertible data from which to reason." — (Ut sup., p. 341.)

This passage places in a clear light the conditions of the
problem and the difficulties attending it. The author does
not, however, propose a solution of the difficulties, nor does
he enter into any theoretical discussion. His account of the

GLACIAL ACTION. 41

travelled blocks is the only one which we have seen ; no other
writer on Arran, that we know of, has turned his attention
to the subject. It is hoped, therefore, that the notices now
given will be the more acceptable.

In Arran, as generally in other districts, the boulders be-
long to a particular period. The entire system of rocky
strata had been formed, and the existing inequalities of the
surface established; but in all probability the last upward
movement of the land, to which we have already often re-
ferred, had not taken place. The relative age, in fact, seems
to have coincided with that of the boulder-clay of Scotland,
or with the newer pleiocene era. Then, as regards the forces
concerned, we know only two natural agents capable of pro-
ducing the effects. These are currents of water and moving
masses of ice. Now, the former are totally inadequate to
carry forward masses of the enormous magnitude found here,
or even to transport the lesser blocks over all the obstacles
which they have surmounted, in their outward course from
the parent rock. Besides, they are often found " perched"
in situations where it is extremely improbable that currents
could have left them, and also crowded together in groups in
places quite open, and removed from the influence of eddies.
It is true, indeed, that the origiff of such currents can be
readily accounted for, by movements which we know to have
taken place — the elevation, namely, of the mountain nucleus
from beneath the sea. We have only to suppose that it was
sudden and of considerable amount, and we have at once
generated a series of mighty pulses, which would carry the
disturbed waters, with their load of torn off materials, along
the surface of the lower lands still submerged. Rocky ma-
terials may thus have been swept away and re-arranged in
new situations, valleys scooped out, and extensive denuda-
tion effected. But the forces thus brought into play cannot
have been adequate to bear along the enormous masses, now
far separated from the parent rock ; and therefore we do not
hesitate, on this and the other grounds above stated, to con-

42 GEOLOGY OF ARRAN.

elude that moving masses of ice were the tran spoiling agents,
la the passage of glacier ice adown the valleys, and the
buoyancy of floating bergs, forces of sufficient energy would
be lodged to carry the largest masses ; and this agency we
know is adequate to produce the various phenomena of trans-
port, grouping, and " perched blocks."

20. We seem thus shut up to the conclusion that the agent
in this transport was ice in motion. Now, this agency may
have been brought into play in two ways. The northern
mountains may have formed a mass of ice-covered land, with
glaciers descending to the bold shores of the sea of that
period, while the southern plateau may have been under
water. From the extremities of the glaciers masses of ice,
of which some must have been considerable bergs, and the
sea therefore not very shallow, would float away, carrying
the granite blocks which had fallen on the ice towards the
heads of the valleys, and been borne along on the glacier.
Stranding or melting, these floating masses would throw
down their load of blocks, and thus the shores of the island
and the surface of the southern plateau may have become
encumbered with vast multitudes of granite blocks, chiefly
of that coarse-grained variety which constitutes all the
highest mountains. But Ih another way ice may have been
the agent of the transport in question. The whole island,
elevated in both divisions high above the waters, may have
been wrapped in sheets of ice, across which the granite
blocks, as they dropped from the high peaks and precipices
on which the snow could not rest, or were torn off by the
pressure of the ice from the sides of the glens, would be
carried onwards in all directions by the slowly descending
viscous mass. Existing glaciers do not require very steep
slopes for accomplishing the transport of large masses. A
moderate inclination is sufficient; and we need not therefore
suppose that there existed any very great difference in the
relative levels of the surface in Arran from that which now
obtains. All the largest blocks are found at the lowest

GLACIAL ACTION. 43

levels, and comparatively near the granite nucleus — the lesser
may have found their way across the surface of the southern
plateau, the valleys being all filled with ice, borne along with
the slowly advancing mass. Our views formerly inclined to
the iceberg theory and submarine deposit of the blocks ; but
a recent careful examination of the lake district of England,
and a comparison of the markings with those of Switzerland,
have led us to the conclusion that the agency of floating
bergs is insufficient to have produced the regularity and per-
sistency which the markings and other evidences of ice
action now present, and that an icy envelope in a state of
constant advance will alone explain them. This same force
is sufficient greatly to alter the forms and dimensions of
valleys, though not, in our view, to scoop them out in the
first instance ; while it is generally allowed that on this
view we best account for the great detrital accumulations
at the mouths and along the sides of the mountain glens.
This supposed state of the surface of the land implies
the existence of temperatures in the atmosphere and in
the waters of the adjoining sea, such as would favour
the development of an arctic fauna in the waters — a con-
dition of things, indeed, like that now prevailing in parts
of Greenland, where the ice, which covers the land, sends
down glaciers to the sea level; and underneath the rim of
ice, which fringes the coast, a peculiar group of testacea
flourish, very different from those of the British seas. Now,
such an assemblage of arctic species of testacea actually
exists in many parts of the basin of the Clyde, and along the
shores of its friths, and has very recently been found in Arran
also, but elevated to various heights, sometimes sevei'al hun-
dred feet above the pi-esent level of the sea, — the species being
now extinct in these islands, and only known as denizens of
the Greenland and other arctic waters. The conclusion is
perfectly legitimate that, coincident with the disappearance
of the ice, the land sustained a general elevation, which
placed the shelly deposits elaborated beneath the waters high

44 GEOLOGY OF ARRAN.

above their level, and introduced the temperature and general
conditions which, while proving destructive to many species,
favoured the immigration and development of that assem-
blage of species by which the marine fauna is now distin-
guished. But the present level of the shores was not yet
attained, nor the actual coast outlines as yet carved from the
rocky border which broke steeply down all round the island.
The sea covered the plains of Brodick and Shiskin, and
stretched its winding arms far up the solitary glens. During
the slow progress' of perhaps forty centuries, the streams
from the rugged mountain sides and gentler hill-slopes bore
down detritus of granite-sand, slate, and quartz pebbles, and
spread them out below the waters of the quiet friths. In
sheltered places, the tides and waves cut a low but well-
marked margin along the highest water line ; while on the
open shores the heavier surge wore deeper, the hill-slopes
were cut into a steeper and higher cliff, and hollowed out
into caves in all the rocks alike. The testacea and other
denizens of the present shores already inhabited the waters
of that remote period ; but we have no evidence that man
had yet appeared. It is most probable, indeed, that the last
elevation of the land, to which we have already often
alluded, took place before the human period. We are only
certain, however, that all the existing levels were established
prior to the Roman invasion. Here, as generally in the
West of Scotland, this last elevation amounted to about
forty feet, and gave to Arran its present maritime border,
and the inland cliff which forms a singularly picturesque
feature in its coast scenery (see fig. 1, p. 2).

Shell Beds.

21. The occurrence of recent shells of arctic species in the
deposit called the " boulder-clay," shews that in Arran, as on
the mainland, a climate prevailed favourable to the develop-

SHELL BEDS. 45

ment of glaciers. This boulder-clay, or "till," has long been
recognized as existing in Arran ; it is seen in fine section in
the lower seaward portions of many of the gleus, especially
in the south and south-west of the island, and is probably
spread out, though in thinner layers, over much of the sur-
face between the river channels, which are of course best
adapted for the retention and exhibition of detrital matter.
The deposit consists of several beds, with distinct and per-
sistent characters, which are often overlooked, so that the
whole is regarded as one formation. The lowest portion
next the natural rock is an unsti-atified clay, of a red or
chocolate brown colour, in which large boulders, both angular
and rounded, and usually striated, are thrown together pell-
mell. Over this are beds of laminated clay and unstratified
sandy clay, with very few boulders, but occasionally striated
pieces of rock. This is the chief repository of the arctic
shells; but over it are other beds of a character somewhat
different, among which occur shell beds with species such as
now inhabit our own shores.

Mr. Smith, of Jordanhill, was the first (Mem. Wern. Soc.,
Jan. 26, 1839) to distinguish between the shelly deposits of
the West of Scotland — to assign them to two distinct ages —
to recognize the arctic character of the lower, and thence to
infer that a sub-arctic climate once prevailed in this country,
connecting with that of the shells other evidences of glacial
action. This was an important generalization; it at once
introduced order into the heterogeneous mass of beds, ranged
conflicting facts and appearances harmoniously under one
head, and formed a new era in post-tertiary geology, like
that established among the • tertiary beds by the generaliza-
tions of Lyell.

The credit of the first discovery of boreal shells in Arran
is due to the Eev. Eobert Boog Watson, B. A., F.B.S.E., of
Edinburgh, who has kindly furnished the annexed list of the
shells. A detailed account of the beds and their fossil con-
tents was laid before the Royal Society of Edinburgh, and

4G GEOLOGY OF ARRAN.

published in the Transactions of the Society, vol. xxiii. part 3,
p. 523. Soon after we visited the chief locality, which is on
Torlin Water, about one mile inland, and twenty or thirty feet
above the stream, and found seven of the species, in an hour's
digging, besides two species of foramenifera not in the list.
The following is a list of the shells : —

Balanus crenatus.

Purpura Norvegica.

Tellina Baltica, a brackish- water variety of solid ula.

Cyprina Islandica.

Astarte elliptica.

arctica.

compressa.

striata, a very large variety of compressa.

Cryptodon Sarsii.

Modiola modiolus.

Leda pygmsea.

pern ula.

Pecten Islandicus.

opercularis.

Littorina littorea.

Turritella communis.

Natica ?

The perfect state of preservation of most of the shells indi-
cates a quiet deposit on the sea bottom ; the elevation of the
bed above the present sea-level expresses the amount of
former depression below that level ; it expresses, indeed, a
much greater depression, since many of the shells are exclu-
sively deep-water species : the Cyprina Islandica, for ex-
ample, is not met with in our seas higher than at a depth of
ten fathoms. Further, it follows from what has been stated
in regard to the relative position of the beds, that if we can
trace an undoubted boulder-clay deposit up the glens to the
height of 1000 or 1500 feet, even if we find no shells, the
elevation unquestionably indicates this amount at least of
former depression.

SHELL BEDS. 47

22. The arctic or sub-arctic climate which prevailed in the
circumstances we are considering would cover the hill-tops
with perpetual snow; glaciers would fill the heads of the
valleys, then greatly reduced in length by the submergence
of the lower lands, and the marine fauna of the period would
flourish under and outside the rim of ice which girt the
island ; while bergs and floes would float away into the open
water, bearing up buoyant for a time, and then throwing
down their load of blocks to encumber the southern plateau,
as already pointed out (Art. 20). The disturbances attendant
on the re-elevation of the land could not fail to produce a
certain re-arrangement of the surface beds, and even occa-
sional intermixture, and the heaping up of detritus in peculiar
forms, such as those already noticed in the case of Glen lorsa.
The boulder-clay, by its position, would in many cases be
protected from such action, while in some situations more
exposed its upper surface would be subjected to great erosion
or removal, — and hence, no doubt in part, the varying depth
and undulating surface of this deposit. Its original forma-
tion was most probably contemporaneous with the incipient
depression of the land, and due to the joint action of ice and
water ; while its internal structure, huge boulders, unstrati-
fied character, and the almost total absence of shells, indicate
a period of violent disturbance, a hurried pell-mell admixture
and deposit of earth and rocky materials. The angular form
of many of the blocks, the perfect and very general striation
of the angular and rounded alike, so that in many places
non-striation is the exception, cleai-ly shew that, by whatever
agency this singular deposit was formed, and the striation
effected, there cannot have been afterwards a lengthened
transport by rivers, as this is known in the case of the
Alpine streams to produce a speedy obliteration of such
markings.

Further details regarding this curious deposit will be given
in one of our Excursions, and the subject of the shell beds
and the order of superposition of the clays treated at greater

48 GEOLOGY OF ARRAN.

length. Shells from the later deposits, the upper portions of
these beds, and from the surface of the marine terrace and
caves of the old sea cliff, have been found in a few places.
These all belong to species now inhabiting our seas; they
indicate that our climate had changed its sub-arctic character,
and become similar to that which now prevails.

49

EXCURSIONS IN AEEAN.

EXCURSION I.

TO THE SUMMIT OF CIOR-MHOR.

23. THIS shall be our first excursion. It will reveal to us
tlie structure of the granite nucleus, and the relations of the
mountain groups. It will shew us the basset edges of all the
strata as we pass in succession across them to the central
granite. We shall see some curious dikes, explore a lonely
corrie, climb a high pass, and thread a difficult pathway by
the edges of the highest cols. The morning clouds have
melted from the peaks; there is neither bank of mist nor

cirrcms haze to veil the far-off horizon ; the day will be calm
and bright, and the view from the summit glorious. Cior-
Mhor (Kior-Yawr) is that far-off peak, with sharp point and
rugged shoulders, rising behind the plateau on the south-
west of Goatfell. We shall reach it by a walk longer and

E

50 GEOLOGY OF ARRAN.

more difficult than the ascent of Goatfell, and the more ex-
citing that it is somewhat dangerous. Should a mist happen
to surprise us upon certain portions of the route, our situa-
tion might be very critical. How promising soever the day
may be, let the climber never enter these mountains without
a pocket compass. The mists come down so sudden and so
thick in this changeful climate that without it he will be
bewildered and lost amid the high cols and peaks, and huge
slippery granite sheets. With ordinary caution, and the use
of a correct compass, he has nothing to fear. True, he will
find neither house nor herd's shieling within the mountain
circuit; but directing his steps by the compass, he cannot
fail, in a walk of two or three hours at the utmost, to reach
the inhabited border, where a frank welcome will meet him
at every cottage door. But let him bear in mind that the
variation of the compass in Arran is now about 26J0 west of
true north, with a yearly decrease of 4' or 5'. To the geolo-
gist we would recommend to carry, besides his hammer,
indispensable at every step, a good clinometer with which to
note the varying dips and inclinations of the slate and sand-
stone, and the relations of these to the granite centre and
the numerous dikes. To the botanist, besides his vasculuin,
a field book will be very useful, for immediate pressure of
the delicate mountain plants. On this and most of our
excursions our departure will be taken from the shores of
Brodick Bay, the unrivalled grandeur and beauty of which
attract the greatest number of summer visitants. For a few
days' sojourn new inducements are now afforded by a well-
managed and spacious hotel, on a beautiful site at Invercloy, on
the south side of the bay. Here, and at Corrie and Lamlash,
steamers call several times daily during the summer months.
24. The rock on the shore at Invercloy is a conglomerate
of the age of the coal — a member, in fact, of the coal forma-
tion. It was formerly classed as a lower member of the
New Red, but this view cannot now be maintained. The
inland cliff marking the old coast line is well seen on the

EXCURSION I. 51

Invercloy shore, and extends far up both sides of Glen Cloy.
The lower part of the glen, much of the plain of Brodick,
and the marshy grounds at the head of the bay, are but an
expansion of the terrace which formed the sea bottom when
the tides and waves were carving out the cliff. The alluvium
and rolled stones form, however, but a thin covering to the
subjacent sandstone, which appears in the river bed a little
way up the glen. It appears also in the bed of the Rosa
burn, but nowhere in the Brodick plain. This plain, indeed,
is but an old expansion of the beach. The bay formerly
had its termination some way up Glen Rosa. The mound
in front of the glen, on which a farm-house lately stood, is
but forty feet above high water ; the mouth of Glen Rosa is
much less; in fact, it is quite within the limits, to a con-
siderable distance up, of the old sea terrace. Brodick plain
shews beneath the soil a continuous covering of rolled gravel
cemented by iron, and impervious to water. To drain it in
any part, it is only necessary to pierce this covering, when
the water at once disappears. On the north-west of Glen
Cloy the sandstone is quarried upon the line of cliff, and
forms a tolerable building stone. The dip is nearly south,
at about 25°. By the side of Brodick wood, adjoining the
neat school-house erected by the late Duke of Hamilton,
a vein, or perhaps bed, of pitchstone occurs in the sand-
stone. A portion only of the front appears by the side
of a lane, shewing a prismatic structure in the rock, and an
underlie towards the west. In large blocks lying loose upon
the surface, a similar structure is seen. The colour is bottle-
green, and specks of red felspar disseminated in the base give
the rock a porphyritic texture, approaching that of pitch-
stone porphyry. The direction is nearly south-east and
north-west. It is probably a continuation of one of the
many beds or veins of the Corriegills shore, which we shall
notice in another Excursion. A breadth or width of fully
thirty yards is exposed ; but it is very difficult to say whether
it is a bed or a vein. We were unable to trace it north-

52 GEOLOGY OF ARRAN.

wards of this point; it seems to be cut off by a dike of dis-
integrating ironshot greenstone, forming a bank behind the
school-house, which has an angular course with respect to
the vein. In the opposite direction the dike intersects the in-
clined strata of sandstone, but no marked change is produced.
It is in excellent taste that the striking geological features of
this spot have been left untouched in carrying out the improve-
ments connected with the erection of the school-house.

By the wayside here, where a sweep of the road gradually
opens to us one of the finest views in Arran, taking in the
first great reach of Glen Rosa and its magnificent background
of mountains, there stands a huge upright stone, marking,

Fig. 13.

perhaps, the spot where a chief was interred, or where a
leader fell in the old days of feud and warfare ; or mayhap
the scene of some decisive battle with the old Norse invaders.
Many such stones are found in the island ; but their purpose
and date of erection are wholly matters of conjecture. There
are several in this immediate neighbourhood; on the high
ground south of Invercloy, and on the plain of Glen Shant,
between the mouth of Glen Rosa and the site of the old
village of Brodick. A complete circle of such stones formerly
existed at the mouth of Glen Shirag : it is briefly noticed
in Headrick's Arran; but not a vestige now remains; in
1813 the stones were broken up and removed, to make way
for the operations of the plough. In most places where we
examined these stones, we found them to be coarse sandstone

EXCURSION I. 53

of the old red formation. As this does not exist in many of
the localities where the stones now stand, we must conclude
that mechanical appliances of great power were brought to
bear in their transport; and therefore it is not wonderful
that, in a rude age, their erection was ascribed to the hands
of giants. In all ages the illiterate observe facts and pheno-
mena with tolerable accuracy ; but their explanations always
introduce the marvellous or the supernatural. The subject
will be again referred to.

25. There are two paths to the entrance of Glen Rosa; we
take that which passes Brodick Church, and crosses the
opening of Glen Shirag. The church stands on a platform,
bounded northwards by a low cliff of sandstone, and over-
looking one of the most varied and pleasing views in Arran.
In this sandstone, underneath the north wall of the church
enclosure, there is a bed of carboniferous limestone in a ver-
tical position. It has been largely quarried, and a small
portion only is now visible. It contains fossils, of which the
most characteristic is the productus giganteus, completely
identifying this bed with the limestones of Corrie, the Salt
Pans, and Ben-leister Glen'; and enabling us, therefore, to
assign the sandstone also, without hesitation, to the age of the
coal formation. A little farther on, above a rustic bridge,
where the Shirag burn, rushing out from a winding rocky
gorge overhung with trees, forms a scene strikingly pic-
turesque, another bed of limestone occurs in the sandstone.
Thence to the entrance of Glen Rosa, we pass across the lower
beds of the carboniferous formation, which, however, are no-
where seen except in spots in the bed of the Rosa burn.

We now reach the outer edge of the band of old red sand-
stone, which, ranging from the Corrie shore diagonally by
the flank of Goatfell, crosses the opening of the glen. It is
seen on rocky prominences by the side of the path, but no
junction is visible. The beds here exposed consist of a dark-
coloured, close-grained sandstone, with specks of mica, bits of
quartz, and small clay-galls, and are very characteristic of the

54 GEOLOGY OF ARRAN.

upper portions of the formation. The lower portions are seen
farther up the glen, but at some height on either side, in the
wood and moor on the left, and the hill-side on the right. They
are very coarsely conglomerate. Specimens of both varieties
may be seen in the stone fence by the side of the wood.

rf

Fig. 14.

(a a) Sandstone and conglomerate ; (b) productus limestone ;
(c) old red sandstone ; (d) schist ; (e) granite.

The alluvial mound at the entrance of the glen has been
noticed already, as most probably the terminal moraine of a
glacier which once filled the valley. It is precisely in the
position where such a moraine would have been thrown
down, and consists of such materials as the ice would have
borne forward ; and its height places it far above any existing
river action. It appears to have extended, at some former
period, entirely across the opening of the valley, backing against
the hill side on the south, as it now does in the opposite direc-
tion, and forming the barrier which confined a lake, occupying
at that period the lower portion of the glen. The bursting
of such barriers, and their subsequent modification by floods
in the river, are common phenomena in mountainous districts.
Traces of a lateral moraine are seen on both sides; and
farther up the glen, where it turns northwards, two other
mounds, rising high above the stream, are in the position
where a terminal moraine would be thrown down, after the
ice had retreated from the lower part of the valley.

The discovery of an anchor in Glen Rosa, similar to those
now used by the herring smacks which visit Brodick Bay,
has often been referred to as proving that since the island
was inhabited by a people far advanced in civilization, the
sea filled the valley, and afforded a " trustworthy station for
ships." That an anchor was really found, brought to the

EXCURSION I. 55

smithy at Invercloy, and worked up into vaiious articles, we
think there cannot be a doubt, from the accounts given, by
several persons still living, who saw it and handled it. All
agree, howevei1, in fixing the locality in which it was found
at a place where peat is cut, at a considerable height on the
southern hill side, and therefore far above the level at which
the waters of the sea stood before that last elevation of the
land to which we have so often alluded already. If the
discovery, then, be admitted as a fact by the archaeologist, it
is entirely without that geological significancy which attaches
to the canoes found in the holms on the banks of the Clyde
and within the city of Glasgow, which clearly shew great
changes of level.

26. The contact of the lower old conglomerate with the
clay slate is not seen in Glen Rosa. The latter rock first
appears in the bed of the stream, at the sharp turn where
it begins to flow eastwards; but the junction must be
farther down the burn. The slate rises high into the hills
on both sides, forming on the north the principal mass of
Glen Shant rock, called also the Pillar, from a large isolated
sheet or prism, standing out detached from the front of the
bold cliff". The precipice is about 1100 feet above the river,
and forms one of the finest features of this noble glen.

We now approach the base of the series, where the central
granite rises from beneath the enveloping slate rocks; and
here a celebrated junction occurs in the bed of the stream.
The hill sides shew the contrast of the two rocks from a
great distance, in the bare stony character and loose gray
blocks on one part, and the grassy or heath-covered slopes,
with dark terraced ledges, on the other; and the geologist is
therefore prepared to find a junction somewhere here in such
a natural section as the river affords. It occurs about two
or three dozen yards below the point where the Rosa burn
receives from the west side its only tributary — the Garbh-Alt
or Rough burn, which drains the whole eastern side of the
Ben-Ghnuis range, and comes down into the glen, bounding

56 GEOLOGY OF ARRAN.

headlong across the huge granite sheets in a series of striking
falls. The junction has not been so well shewn these few
Tears as it was formerly, iu consequence of the accumulation
of loose masses of granite; still it is sufficiently well seen to
make the place interesting and instructive in a high degree,
and some future floods in the river may again open it better up.
The slate is greatly altered for a considerable distance down
the stream, and pervaded by small veins and strings of quartz,
and granite in which quartz predominates. The colour of
the slate is changed, having more of blue than is usual to it ;
the structure is altered also, the laminae are contorted, and
present thin bands of different colours, chiefly blue and gray;
the latter being purely siliceous, that is, flinty slate or
quartzite without the colouring matter, iron or manganese,
which exists in the former. The slate on this side of the
mountains is generally a dark-coloured, coarse, siliceous rock,
shewing shining crystalline flakes in fresh fractures; in some
places assuming that arenaceous, semi-conglomerate structure
which used to be designated as greywacke. Both structures
are obliterated on approaching the granite; the colour is
bluish, or blue and gray in alternate bands, the structure is
extremely fine-grained, and the hardness and toughness are
both excessive. These changes, the contortion of the laminae,
or the total disappearance of all stratification, coupled with
other modifications not seen here, but to be again noticed,
clearly indicate that the schist to a considerable distance
from the granite was subjected to intense heat, and remained
in such a state of at least semi-fusion as to permit, under the
action of chemical forces, a new arrangement of parts, and
the permeation and interlacing of veins from the fluid
mass below. The granite veins are less numerous, smaller,
and of varieties differing from the ordinary type of this rock
more than is usual in most other junctions with the slate ;
such might perhaps be seen in the interval of several yards
here obscured before the granite itself is reached. The great
extent of the altered slate seems to indicate that the strata

EXCURSION I. 57

near the junction are of inconsiderable thickness, and that
the granite exists beneath at a small depth. The slate is
here traversed by a basaltic dike, still further modifying its
altered structure. The dike intersects the bed of the stream
at a small angle, but is seen only on the east bank; it ranges
about magnetic north and south, and has a breadth of ten
or fifteen feet. We shall meet with it farther up the glen
on another excursion. Rising from the bed of the stream in
great rhomboidal masses, divided by partings here inclined
towards the slate, the granite extends in a slanting direction
up the hill towards Goatfell, and on the opposite side by the
Garbh-Alt burn, so that the line of junction almost coincides
with the southern margin of the stream. The remaining part
of our walk is entirely on the granite.

27. Our path now lies up the steep slope forming the
western side of the glen, a little to the north of the Garbh-
Alt, which we keep on our left hand. Having reached the
summit level we are on the southern slope of the high
rugged ridge called Ben-Cliabhain, which forms the western
boundary of Glen Rosa. In the hollow between it and the
lower swells of the Ben-Ghnuis range, the Garbh-Alt pursues
its rapid course from north to south, along a granite bed,
down a pretty fall, and then between perpendicular walls of
granite, about twelve feet in height, till, escaping from this
rocky barrier, it sweeps round the south end of the ridge,
and plunges headlong into the depths of Glen Rosa. The
gorge has been excavated along the line of a basaltic dike,
which occupies the bottom of the stream throughout, and
retires from it at the base of the fall. These dikes are
prismatic across; and this structure renders their disintegra-
tion much more easy than that of the granite. The amount
of wearing in this case is measured by the depth of the chasm
and the distance to which the fall has receded. The stream
ran at first on the level of the top of the granite walls, and
the fall must have been at the southern extremity of the
chasm ; the recession would cease, or become extremely slow,

58 GEOLOGY OF ARRAN.

when the present situation of the fall was reached, as the
dike here retires from the stream. The dike is ten to fifteen
feet wide, and ranges nearly due north and south. The rocks
adhere firmly at the junction, but the alteration on the
granite is not remarkable. Similar cases of the excavation
of river channels along the line of dikes are frequent in
Arran. As a general rule, fractures or faults determine the
course of streams in the first instance; along such lines the
excavation is much more rapidly effected.

The steep brow on the eastern side of the wide hollow
where we now are, exhibits many rounded masses of granite,
presenting the "moutonneV character of surface, as if
moulded by the action of ice ; but as no striae were observed,
we can hardly ascribe them to the action of glaciers, as the
forms may be due to the effects of disintegration on the con-
centric structure upon the large scale, so often seen in
granite. The bed of the stream, as we pass up, is strewed
with many loose rounded masses of pitchstone and trap,
indicating the existence of dikes or beds of these rocks
among the lofty precipices on the west. Mounting this
steep brow, to reach the corrie under the north front of Ben-
Ghnuis, we meet with a dike of spheroidal trap in the bed
of one of the streams ; it is about seven feet wide, and ranges
35° W. of N. The rounded masses of granite here may
have received their forms from the long-continued action
of water trickling over them, and torrents occasionally
sweeping along gravel and large stones. Arrived at this
corrie, we are in the midst of a scene wild, lonely, and
picturesque. The bare and rugged precipices of Beu-Ghnuis
rise high into middle air on the south, with their immense
sheets and rhombic masses of granite, from six to twelve feet
in the side, piled up, block on block, in massive courses, like
the huge rough masonry of giants. The topmost row, broken
by clefts and deep gashes, due to irregular disintegration,
shews grandly in its perfect definition against the clear sky.
Along the front, which sweeps round to form one side of the

EXCURSION I. 59

corrie, there stand out here and there enormous pillars of the
rock, detached from the cliff behind, resting on a basis
which is rapidly giving way under the active agents of waste
in this changeful climate, and threatening a speedy descent
into the valley of the Garbh-Alt. The rugged outlines of
the Goatfell group bound our view on the east, the distant
landscape being shut out by the intervening ridges ; and the
eye from this point does not take in a single human dwelling,
or other sign of the abode of man. No sound reaches the
ear but that of the crystal rills trickling from the clefts of
the granite, the hum of insects on the wing, or the twitter
of the solitary stone-chat, as it flits from rock to rock. The
solitude is complete, the silence solemn and impressive. Our
perfect isolation amid such a scene — the vast dimensions of
the objects ai'ound us, and their expression of power, are
true elements of the sublime, and awaken the most pleasing
and elevating emotions. There is a delightful consciousness
of a new activity in the fancy, and an increased buoyancy
and intensity in the feelings. To the geologist there is
another source of the sublime in contemplating the effects of
the mighty forces which have rent the crust of the earth,
raised these mountain masses from the fiery depths beneath,
and scooped the glens and corries out of the solid rock.

28. The cliff on the north side of the corrie shews some in-
teresting dikes. One of these is of green pitchstone, and cuts
the granite sheer through in a north and south direction
from bottom to top of the cliff. It is four feet wide, prismatic
across, and, owing to the more rapid disintegration, depressed
below the level of the granite. The contact does not present
any peculiar change in either rock, such as usually marks
the plane of contact of pitchstone and the sedimentary strata.
The pitchstone is decomposed into a thin white film in many
places along the outer edge of the dike, next the granite, in
consequence, probably, of the oxidation and removal of the
iron which enters into its composition. The dike is in some
parts of its course obscured by debris, but upon the whole is,

GO GEOLOGY OF ARRAN.

perhaps, the best defined dike of this rock occurring any-
where in the granite of Arran.

Two basaltic dikes occur close together, about 100 yards
east of the dike we have been describing; they traverse the
granite precipice in the same manner, but in a different
direction, their course being about N. 28° W. (true), subject,
however, to undulations. These dikes are from eighteen
inches to two feet broad, and are separated by a granite band
eight or ten feet in breadth ; elliptic masses of granite, of
which the largest we observed was about eighteen inches by
nine, are enclosed in the trap, but very little altered. The
alteration, indeed, is nowhere remarkable, the granite being
in some places coarse, in others fine-grained, along the planes
of contact. Specimens may be obtained of both rocks firmly
adhering. All the appearances point to the forcible intru-
sion of the trap.

Some pretty plants occur here in shady spots on the
granite ledges, where a little soil has accumulated ; they will
reward the young botanist for his long walk to their
secluded habitat. The Sedum rhodiola, Oxyria reniformis,
Saxifraga stdlaris, Alchemilla alpina, and several others,
rejoice in the temperature and humidity which these heights
supply.

To reach the head of the valley, and the first ascent of the
Pass into Glen lorsa, we now direct our steps along the base
of the precipice, where the grassy tufts and granite debris
afford a safe footing on the steep slope, keeping as high a
level as possible, in order to shorten the ascent by which the
summit is reached. As we pass along we notice several
dikes of pitchstone and basalt, ranging north-westwards up
the precipices on our left towards the summit of the ridge of
Ben-Tarsuin (2706 feet in height), and doubtless crossing
down on the other side* into Glen lorsa ; but we may not
now delay to trace them. The lower part of the Ben-
Cliabhain ridge on our right (2141 feet in height), and that
along which we have passed, coalesce at the head of the

EXCURSION I. 61

valley, and their union is marked by a very steep grassy
slope, encumbered with granite blocks projecting from the
soil in situ, or deeply imbedded in it, but free from the huge
flat and smooth sheets, along which it is difficult and dan-
gerous to cross. This depression or break is in the direction
of the head of the valley, and owes its origin to a basaltic
dike, which appears at the beginning of the ascent, and is
seen to enclose masses of granite, as in the case already men-
tioned. Its range is magnetic north and south, and width
about twelve feet. The situation of the Pass of Bealach-an-
Fhir-Bhogha, at the north base, to which we are now to
mount, and which is fully 1000 feet above us, is indicated
by a bold rocky point, a little in advance of Ben-Tarsuin,
2706 feet high, the last high summit of the Ben-Ghnuis
range. Marking its position by the compass, and then
pressing up the steep, we gain a wide and grand prospect
from the summit of the Pass. Clambering to the top
of the rocky point, we look down from a height of fully
2000 feet into a nook or recess of Glen Rosa on one side, and
into Glen lorsa on the other. The descent towards the
latter is easy; towards the former somewhat dangerous.
From the Garbh-Alt valley, in fact, the only safe access to
Glen Rosa is by the way we have come up.

29. We are now at the southern extremity of the Ceims
(Kyims), which link on Cior-Mhor to the Ben-Ghnuis range.
This is the ridge whose sharp and rugged outline, seen from
the shores of Brodick Bay, is well known as bearing a strik-
ing resemblance to the profile of a late distinguished states-
man and writer. The ridge is formed by the edges of vast
tabular masses or sheets of granite, inclined towards Glen
lorsa at a considerable angle, and cut sharply down on the
side next Glen Rosa, so as to present towards it a continued
precipice, formed of successive tiers of granite sheets and
rhombic blocks. The jagged outline is due in part to the
irregular wearing of the coarse-grained granite, but still
more to the intersection of the ridge by a series of whin

62 GEOLOGY OF ARRAX.

dikes. The horizontally prismatic structure of these sub-
jects them to a more rapid decay than even the friable
coarse granite ; and hence most of the deep notches of this
jagged ridge mark the situations of whin dikes. The fact is
curious and interesting, and has not been before noticed.
Ranging up the front of the precipice from Glen Rosa, these
dikes cut right through the crest of the ridge, and pass down
towards the fine-grained granite, which occupies most of the
lorsa valley, and rises up on the back of the Beu-Ghnuis
range and the Ceims as far as the level of many of the cols
— that is, the lowest parts of the ridges between the glens.
There can be no doubt, from the direction in which these
and the other dikes already mentioned range, that they are
the same as those which are found crossing the fine-grained
granite tract on the west side of the lorsa valley, and will be
mentioned again.

30. The jagged and notched character of the ridge makes
it impossible for us to pass along it on our way to Cior-
Mhor ; neither can we safely cross the huge granite slabs at
the back of the ridge, as they are smooth, slippery, and con-
siderably inclined. But below these a safe, though rough
and irregular, pathway will be found ; and from this we can
occasionally pass upwards towards the ridge, along the clefts
in which the dikes lie, to have a peep down into Glen Rosa, or
to scan the frowning cliffs on the north side of Ben-Tarsuin.
For the latter purpose a telescope will be useful. We must
be careful, however, to return by the cleft by which we came
up, till we reach the path. The granite sheets must not be
attempted, as a single false step upon them might hurl the
climber with fearful velocity into the valley at their base,
filled with blocks and debris. The range, width, and struc-
ture of the dikes are well seen as we pass along ; the width
is various, in some five or six feet only, in others ten to
twelve, and fifteen to eighteen feet. The broadest, being of
this latter width, is one on the south side of the col, between
Glen Rosa and Glen lorsa. It consists of a crumbling green-

EXCURSION I. 63

stone, and ranges N. 29° W. The others have various
ranges between west and north. Many small shining flakes
of crystalline oxide of iron occur in the trap of these dikes.
The ridge of the Ceims is composed of the coarse-grained
granite; but the path at its west base, along which we have
come, is partly on the fine-grained variety. This appears
generally at the height of about 1600 feet on the cols, and on
the west side of the ridge, but is not seen in Glen Eosa nor
on Goatfell. It disappears on the ascent of Cior-Mhor, and is
succeeded by the coarse-grained variety. The contrast is
remarkable. The rock has quite a different aspect, a dif-
ferent feel under the hammer, and a peculiar style of disin-
tegration, giving smooth outlines, and an entire absence of
the aiguille-like highly picturesque forms, into which the
coarse variety is resolved by the action of the atmosphere.
The decomposition of both rocks conceals the contact, and
we were not so fortunate as to discover anywhere here the
actual junction of the two varieties. But the day is waning,
and we have yet to scale the lofty peak of Cior-Mhor, shoot-
ing grandly up 900 feet above the ridge on which we stand.
Thoxigh right to the summit " we might press, and not a sigh
our toil confess," we must pause now and again to mark the
ever-changing features of the magnificent scene gradually
opening towards the west and north, and the new aspects in
which the rugged crest of the Ben-Ghnuis range now appears.
We must note, too, the change in the rocky floor over which
we are passing. We leave the fine-grained granite on gain-
ing the foot of the steep ascent; thence to the summit the
mountain is composed wholly of the coarser kind. It is dis-
posed in irregular tabular masses, split up into rhombic or
cuboidal forms by fissures, independent of disintegration, and
coeval with the solidification of the rock. The thinner
masses we have called sheets; in both, the divisional planes
separating mass from mass, and the fissures perpendicular to
them, are alike the result of crystallization on the large scale,
and bear no analogy to stratification, which is the result of

G4 GEOLOGY OF ARRAN.

sedimentary deposit. The disintegration of granite, porphyry,
and other igneous rocks, is mainly determined by these lines
of separation; in some granites, but more remarkably in
porphyry and the trap rocks, by a concretionary structure,
which has resulted from the mode in which the crystalline
centres of affinity develop themselves at the first parting of
the heat of fusion, in a melted mass beginning to solidify.
The schistose and prismatic forms, under which granite often
appears, are but slight modifications of the forms already
noticed, depending on the relative position of the divisional
planes. The schistose form has often been described as a
true stratification ; but this structure is not continuous in
one direction as strata are, nor does it exhibit the fracture or
incurvation of beds ; it is, in truth, but a local modification
of the rhombic or cuboid form, under which granite more
frequently appears.

31. The summit of Cior-Mhor, narrowed by distance into
the form of an alpine aiguille, is found to be an irregular
elongated plateau, large enough to accommodate a small
pic-nic party. The nigged shoulders flanking the peak are
huge rifted masses of bare rock, separated by clefts which
descend far into the heart of the mountain. On three sides
there are precipices, through clefts in some of which the top
may be reached ; an easy ascent is possible from the west
side only. Thrown forward on a salient angle of the western
ridge, and little more than 300 feet lower than Goatfell,
Cior-Mhor affords a commanding view of the ridges and
dividing valleys, the peaks and precipices of this singular
mountain group. Its situation, as the geographical centre of
the tract, has been noticed already, and the relations of the
various ridges pointed out (Art. 2). Viewed from the
summit, where we now stand, the scene is very wild and
grand. The ridges swell up steeply and nobly in front of
us from the very depths of the glens, in their majestic forms
of " peril and pride," and stretch away on either hand,
shooting up here and there into the highest peaks, and cut,

EXCURSION I. 65

in lower parts between, into spiry fantastic crests. The
craggy precipices and long steep fronts of naked rock have
an imposing expression of sternness and power. Crowning
the ridge of which they form the lateral supports, Goatfell
presents on this side its grandest aspect. The eye, from its
elevation, takes in, under a large visual angle, the entire
western steep, from the summit of the mountain to the
bottom of the glens. On the north side of Glen Sannox the
ridge of the Castles and Suidhe-Fergus starts up with little
less of suddenness and grandeur. Lying deep down at the
foot of these lofty ridges, and closed in on the south-west by
the high mountain on which we stand, Glen Sannox has an
air of singular loneliness and solemnity. The same breadth
of form and grand scale of parts are found nowhere else in
Arraii; — the uniform and simple structure of the glen is its
greatest charm. The silver thread of its river, meandering
far out eastwards, leads to the world of life without ; and the
murmur of distant waters, stealing up from its sombre depths,
breaks pleasingly the awful stillness of the summer day on
these high peaks.

32. We have attempted, in preceding paragraphs, to indi-
cate the successive steps of the process by which the moun-
tain nucleus acquired its actual conformation. From our
present commanding position we are better able to estimate
the amount of elevatory force required to raise the high
peaks and massive ridges around ; and the length of time
and intensity of erosive agents which the formation of the
long and deep chasm dividing the ranges demands. Fill up
this chasm and the other glens with solid granite masses, to
the level of the peaks and ridges, over all throw a mantle of
slate, continuous with the present circular boundary around
the nucleus, depress the area full three thousand feet, till the
ocean flows freely over all, and some measure will be ob-
tained both of the force and of the time through which the
present aspects were assumed.

33. The sunbeam is the joy of this mountain wilderness.

F

66 GEOLOGY OF ARRAN.

It lights up the solemn old rocks till they laugh into beauty
under its bright spell. Though devoid of vegetation, which
might throw bright tints around the rugged surfaces, — for
the saxifrage and alchemilla, the cryptogramma and other
ferns, the club moss and juniper, nestle in shady clefts, and
small patches of grass occur only here and there among the
blocks, — yet are not these bare rocky masses without a cer-
tain natural adaptation to produce warm harmonious colour-
ing. The three ingredients of granite have peculiar shades
and different reflective powers ; oxide of iron, always present
as a constituent, passes in decomposing through various rich
tints, and the rocky surfaces themselves, smooth or rough,
dry or moist, are often dotted with small lichens. The
result is a sober but pleasing tint in keeping with the
general expression of the mountain scenery ; it runs through
various shades of gray, purple, and a tempered red or orange.
The effects, seen from Oior-Mhor, are finest in the afternoon.
Marvellous contrasts now lie athwart the stony ridges and
deep glens, adding a wondrous charm to the scenery. In the
depths of Glen Rosa the sun has gone down an hour since,
and a deep gloom has settled on the dark recesses of Glen
Sannox. Sharp shadows of the western ridge, shewing a
perfect profile of its jagged crest, are slowly creeping up the
western front of Goatfell, whose summit is bathed in a flood
of glorious orange light. Thrown back from rock to rock
in mellowed and harmonious tints, it maintains a bright
twilight along the base of the western ridge, by which we
must descend, and throughout the upper part of Glen Rosa,
along which our after path will lie. Taking a last survey of
the surrounding peaks in their gorgeous evening tints, and
contrasting the bold rocky foregrounds, now flooded with
light, with the smoother and fading outlines of the lower
hills, we must hasten downwards. The far off landscape we
shall see better another day from the summit of Goatfell.

The low ridge or col connecting the base of Cior-Mhor
with the next height to the south breaks down steeply, but

EXCURSION I. 67

without precipices, towards Glen Rosa. We can descend
easily at almost any part of it. A few minutes will bring
us to the junction of the two burns, over the most rough and
toilsome part of the long walk yet before us. Farther our
way lies nearly by the side of the Rosa burn, through the
moss and heather, till we reach the Garbh-Alt, where our
ascent began in the morning. The smooth carpet of the
glen is now beneath our feet, and we dismiss all fears of
adders in the path, which troubled us a little in the uncer-
tain light of the last hour. Emerging from the glen, and
crossing the pretty burn by the rustic bridge, we gain the
Brodick side and Glen Shant Lane. In twenty minutes
more we are at the door of our comfortable inn, and are soon
seated at the welcome evening meal. It is pleasant to talk
over the incidents of our long walk : they will awaken joyful
memories on wintry nights, years to come.

68

EXCURSION II.

THE CORRIEGILLS SHORE.

34. TO-DAY we shall stroll leisurely by the sea-side, and
study the eruptive rocks which break through the Corriegills
sandstone. The botany of the shore is rich, and the pools in
the tide-way teem with life. We shall notice in this place
the geological phenomena only. A group of whin dikes marks
the first emergence of rock from beneath the Brodick sands ;
and from this point eastwards the rocky platform exhibits a
complete network of interlacing veins and beds of igneous
products, traversing the sandstone strata. Between the end
of the sandy beach and the landing-pier at Spring-bank there
are several cases of bifurcation of dikes, and of a singular
crumpling of the sandstone strata. The bed of the burn
within the pretty grounds connected with the hotel shews
several greenstone dikes, and the beds of sandstone are well
seen in the banks and rocky ledges over which the water
tumbles. On the shore fine sandstone and conglomerate are
irregularly intermixed, indicating periods of sudden and gra-
dual deposit, and varying forces in the transporting currents.
The sandy strata are red, yellow, and white, and, as we ad-
vance eastwards, predominate over those of conglomerate
structure. The fragments in the conglomerate are mica
slate, like that of Can tire, slate similar to that of the nucleus,
white quartz, and quartz of that peculiar resinous variety
(Art. 7), for which we know no locality in Arran. Pieces of
porphyry also occur, but no fragments of granite. The whole
series dips nearly south, at angles varying from 15° to 20°.
The upper surface of the sandy beds is worn in a singular
way, portions more quartzose, or with a calcareous cement,
standing out in thin, sharp, irregular ridges, while the parts

EXCURSION II. 69

around are worn away, being softer or more ironshot. The
rock has thus a honeycomb structure, like what one often
sees in the worn corallines of the mountain limestone. Nume-
rous bowl-shaped cavities also occur, due, probably, to the
removal of imbedded quartz balls, or the grinding action of
these, by the movements of the waves, when once loosened
from their bed in the sandstone.

Some general remarks have been already made (Art. 17)
on the dikes of this coast, and we shall now only notice the
individual cases of most interest. The whole number is
about sixty, and the direction generally between north-west
and north-east, a few running nearly east and west. Most of
them alter the strata of sandstone more or less. The great
majority are depressed below the level of the sandstone,
owing to the more rapid disintegration depending on
their structure, as already often noticed. They are of all
widths, from eight inches to forty feet, and the sides
are generally parallel, and the course rectilinear or
slightly undulating. Most of them traverse quite across
the rocky platform, and are continued into the cliffs, up
whose front they are seen to range, either level with the sur-
face, forming deep gashes, or projecting like walls. These
cliffs are the old sea margin, and are hollowed into caves along
their bases, and otherwise sea-worn to a considerable height.
The gashes in the cliffs were doubtless formed when the sea
stood higher ; the process is now in great measure arrested in
such situations. But dikes placed under circumstances ex-
actly alike do not waste with the same rapidity. Though the
prismatic structure is the same, the chemical composition
varies, as does also the internal texture, while the adjoining
sandstone varies also in its capability to resist decay. When
the alteration produced 'by the dike is great, the sandstone
will resist disintegration ; if the contrary is the case, the
sandstone may wear rapidly, and the dike project. " From
some experiments made several years ago," says Mr James
Napier (paper quoted in Art. 17), "on the decay of trap

70 GEOLOGY OF ARRAN.

boulders, I found that certain varieties of that rock are rapidly
changed by the action of water ; lime and magnesia being
dissolved out, the iron converted into a peroxide, and a crust
formed on their surface, which is brittle and easily abraded."
Mechanical and chemical differences have thus both to be
considered, as well as the relative powers of resistance of the
dike and the containing rock.

A remarkable group of dikes occurs under the east end of
the high cliffs, near the point where the shore bends south-
wards. One of these is the broadest dike of greenstone on
this coast. Its general breadth is twenty-five feet, but it
widens at one place to forty feet. The sandstone is rendered
very hard and quartzose to the distance of several feet. The
range is 47° W. of N., and the inclination east at a small
angle. A deep fissure marks the course of the dike up the
front of the cliff. This dike is noticed by Playfair, in his
Illustrations of the Huttonian Theory (Works, vol. i. Art.
266), as producing a marked change on the sandstone, and as
indicating the relative durability of the two rocks. A little
east of this dike is another, nine to twelve feet wide, inclined
to the west, and ranging N. 18° E. It offers no remarkable
appearances ; but the next dike east of it, though but seven
feet wide, alters the sandstone more than any other of the
whole series. This is probably owing to the nature of the
rock, which, being a highly crystalline greenstone, must have
passed slowly from a state of fusion. The stratification of
the sandstone is obliterated through a space of seven or eight
feet, and this rock assumes the structure of a claystone. The
case is strongly in favour of the view often advanced, that
the Arran claystones are merely metamorphic sandstones.
Intersecting this dike is another, ten or twelve feet wide,
ranging N. 37° W., dipping E.N.E., and consisting of com-
pact fine-grained greenstone. It is sunk below the level of
the sandstone ; and, on this worn, depressed surface, there
rests a boulder of coarse-grained granite, estimated at about
thirty tons weight. Now, no force of surging waves here,

EXCURSION II. 71

surged they ever so fiercely, could shift the position of such
a mass ; and we must therefore conclude that the huge
boulder now rests where it was originally thrown down by
the glacier or floating berg, which bore it from the granite
nucleus. Perhaps, however, it is not necessary to suppose
that the dike was then excavated to its present level. It is
quite conceivable that as the parts of the dike around the
boulder were worn away, its support may have been loosened,
and so its position may have shifted a little at long intervals.
35. In the sandstone of the cliffs overlooking this point we
were so fortunate as to discover a true carboniferous fossil, a
species of ortlwceras, too imperfect for the identification of
the species, but closely resembling the one figured below (fig.
15), the 0. cinctum of Professor Phillips — the occurrence of
which in this locality is strongly in favour of the view already
advanced (Arts. 8, 24) regarding the age of these sandstones.

Fig. 15.

It may, indeed, be urged that carboniferous fossils pass up-
wards into the lower portion of the Permian (lower new red)
system. But we submit that orthoceras is not a genus of
which this can be said ; a very few cephalopods, allied to

ili GEOLOGY OF ARRAN.

nautilus, being the only animals of this order yet found in
the lower Permian beds.

Still advancing eastwards, we meet with several other dikes.
One of these is depressed ten feet in the tideway, is on a level
with the sandstone along the grassy surface at the base of the
cliffs, and in the front of the cliff is again worn, forming a
deep chasm, which is bounded eastwards by a bold projecting
edge of the cliff, crowned with wood. The dike on the beach,
and fissure aloft, are about fourteen feet wide each, but not
exactly in the same direction, the dike sustaining here a con-
siderable undulation.

36. We now reach the great bed of claystone, the largest
upon this coast, and presenting many interesting appearances.
It forms a vein rather than a bed, as it is placed at an angle
of about 15° with the sandstone strata. The sandstone dips
S. 12° E. at about 15°, while the claystone vein is inclined
in the same direction at an angle of about 30°. The rock is
divided irregularly into prisms by joints perpendicular to the
lower surface of the vein, so that the prisms lean back towards
the south, giving the appearance, when viewed casually, of a
bedding directed northwards. The structure at the upper
surface is often schistose at right angles to the joints, or in the
direction of the vein. The base is of a uniform texture, of
felspathic substance, with quartz pieces imbedded. The
structure varies from a uniform claystone, or clinkstone, to a
small-grained porphyry. The colour is pale yellow, or yellow-
ish-white ; and at the first view the rock might be taken for
a sandstone : it has indeed been described, when seen in its
continuation in the adjoining cliff, as a white columnar sand-
stone (Headrick's Arran, p. 66). The rock on which it rests
is a conglomerate; that beneath which it plunges southwards
is fine-grained sandstone. The upper surface of the vein is
very rough and jagged, with no resemblance to the style of
decomposition among sandstones. The breadth exposed upon
the shore is between thirty and forty yards. Along the level
shore, where the vein rises to the north, its lower surface is

EXCURSION II. 73

exposed iu a wide fissure hollowed out of the sandstone by
the action of the sea. It is here seen to rest upon a vein of
trap, three or four feet thick, and having the same inclination
as the claystone. The line of contact is irregular, and in two
or three places thin bands of conglomerate are interposed be-
tween the trap and the claystone. The lower portions of the
claystone, next the trap, are harder, or converted into horn-
stone ; the conglomerate is much indurated, and assumes the
dark colour of the trap; while the latter becomes a fine basalt?
and is intermixed with the sandstone below, or dispersed
through it in lumps. The posterior origin of the trap vein is
thus clearly indicated. The appearances are correctly de-
scribed by Dr MacCulloch (West. Isles, vol. ii. p. 403), and an
illustrative drawing given (vol. iii. plate xxiv. fig. 1). The
sandstone overlying the claystone along the south side of the
vein is very slightly altered.

Several dikes traverse the sandstone platform between the
claystone vein and the great boulder, some running nearly north
and south, and others neai-ly east and west. The former seem
to shift the latter, producing a change in the direction of more
than 20°. The dip of the sandstone is also affected by these
dikes, being thrown round about 25° towards the west. One
of the dikes, sunk more than two feet below the sandstone,
and eight inches broad, is lost on entering the claystone : it
may be connected with the undei-lying trap vein. Another,
close to the boulder, six feet and a-half broad, consists in the
centre of blue-coloured, rapidly decaying greenstone, and at
the sides of a hard crystalline variety of the same rock, stand-
ing above the level of the central parts and of the adjoining
sandstone.

37. The celebrated Corriegills boulder (fig. 16), under whose
shadow we shall now rest, is of imposing dimensions, and a con-
spicuous object from all parts of this coast. A few on the
Corrie shore exceed it in size ; but they are close on the edge of
the granite nucleus, and we may suppose it quite possible that if
Goatfell "shook his giant sides" under some earthquake throe,

74 GEOLOGY OF ARRAN.

they might have been hurled headlong from the summit to
the sea-level. Other causes must be sought for the transport
of this enormous mass from the parent mountains ; and of
others still farther removed, though of lesser magnitude. We
have already considered the only possible causes, and at-
tempted, to estimate the evidence in favour of each (Arts. 20,

Fig. 16.

22). That to which we chiefly lean receives support from
the case before us. A crowd of lesser blocks- surrounds the
huge boulder of which we speak — an association much more
likely to occur in the case of glaciers or bergs than of currents
emanating from a centre so remote. The cubical contents,
and consequently the weight, are very difficult to estimate on
account of the irregular form. The dimensions at the base
are 21 feet by 12, and the height 15 feet. If rectangular, it
would weigh 315 tons ; but if we deduct one-third for the
conical form, which is a large allowance, we shall have a
weight of 210 tons.

The cliff we have had on our right all along subsides here,
and the cultivated fields of the Dun-an and Corriegills farms
come down to the water's edge. The northern mountains are

EXCURSION II. 75

hidden from us, but the picturesque ridge of Dun-fion sweeps
finely round, abutting on the sea in a lofty cliff. The farm-
houses nestle cozily in against the hill sides, sheltered from
winter storms, and forming a delightful summer retreat for
those who love retirement and quiet simple beauty. The
grander features of Arran can be seen in the distance of an
easy walk.

38. South of the boat station, under the farm-house, the
dikes traverse the sandstone in every possible way ; inter-
secting one another at various angles, bifurcating, lesser ones
lost in larger, &c. One of them is exposed tln-ough a longer
course than any other dike on the coast. It runs a long way
parallel to the line of the shore, or almost due north and
south, till lost under the sea near Clackland Point. It is
fourteen feet wide, sunk under the sandstone in the tideway,
and sends off a branch towards the north-west, a tongue of
altered sandstone being at the bifurcation.

Two pitchstone veins, one of claystone and one of quartz-
iferous porphyry, are found on this part of the shore. The
lesser pitchstone vein traverses the level shore obliquely about
half-way between the boat station and the base of Dun-fion.
Within the tideway it ranges about 72° "W. of N. ; then bends
about 28° towards the south, i.e. — runs about due west, and
bends again into the former course before it enters the sand-
stone cliff. Under high-water mark it is about three feet
wide, in other parts five or six feet. Seaward it is placed
conformably among the sandstone strata, and is irregular in
direction and breadth. A dike of diorite here cuts it nearly
at right angles, but the appearances are in no way remarkable.
The greater pitchstone vein is conspicuoxis, forming a broad
band in the front of the sandstone cliff farther south. It
occupies at first a slanting position in the cliff, parallel to
the sandstone strata, dipping with them towards the
S.S.W. at nearly 30°, and rising towards the north-
west. In the opposite direction, or towards the south-
east it seems to plunge beneath the sea ; but the debris

7G GEOLOGY OF ARRAN.

here obscures it, so that its course canuot be traced
to the water's edge. Climbing up the cliff to examine it
more closely, we find it to be 13 feet 5£ inches in thickness ;
of lamellar structure and dark bottle-green colour. There is
no remarkable change on the sandstone — a slight induration
merely, but the lower portion of the pitchstone is changed
into a blue-coloured, porous, slag-like matter, like a pumice-
ous lava : this, however, is probably the mere result of
decomposition. Its exact position among the sandstone
strata is not easily determined : but £here is little doubt that
it intersects the beds, and is therefore an intrusive vein.
Fallen masses of the pitchstone strew the beach ; and among
these, where the path comes close on the water by the base
of the cliff, there is a dike of red quartziferous porphyry,
nearly perpendicular to the shore, twelve or fourteen feet
broad seawards, but narrowing inland to five or six feet

A granite boulder, about one-third of the size of the one
near Dun-an farm-house, rests here in the tideway ; and great
numbers strew the shore all along.

39. Between the two veins of pitchstone which we have
just described, there occurs upon the flat shore another large
vein of claystone remarkable for a peculiar structure. This is
developed in those parts only which are near the junction
with the sandstone, along tfte south side of the vein. The
appearances are fully and accurately described by Dr.
MacCulloch (vol. iL p. 405). The structure referred to " is
concretionary globular, or striated — the latter being either
found separate or united with the globular in the same
specimen. The former puts on sometimes the appearance of
spots, circular or elliptic, resembling Siberian jasper. The
spots, as well as the stripes, are attended with corresponding
differences of hardness, the former arising from the globular
structure, the latter from a schistose or laminar one. The
spots being often elliptical, compressed, or elongated, occa-
sionally become laminae in the progress of elongation, passing
into them by insensible degrees. The vein next the sand-

EXCURSION II. 77

stone varies much in hardness, but it cannot be said that this
induration bears any relation to its proximity to the sand-
stone. The concretionary structure is seen both in the hard
and soft varieties : it is radiated fibrous, the radii sometimes
diverging from a point and sometimes from a solid nucleus,
which is further, in some instances, surrounded by a white
earthy crust. In the progress of induration the rock at length
loses its character entirely, appearing to pass into a substance
of an indefinable nature, of a horny aspect and dark dull
green colour — partaking of the character both of calcedony
and pitchstone. It has been described as globular pitch-
stone ; but it is far removed from this rock by its extreme
toughness, want of lustre, and by the form of its fracture."
Mineralogists have long regarded this curious rock with much
interest, and various opinions have been held respecting its
true relations — some considering it as allied to claystone, and
others to pitchstone. A comparison of the appearances
observed here with those seen at Tormore, and in Moneymore
glen, has led us to conclude, that " the substance of an in-
definable nature " is homstone ; and a transitional state of
these earthy matters between claystone and pitchstone ; the
globules being due to the formation, within the mass when in
a fluid state, of those crystalline centres of radiation, which
have been observed to exist in melted rock, and which appear
to determine the prismatic and jointed structure when the
fused mass cools slowly. The radiations are of pure quartz,
while the base is of the same substance, but mixed with
colouring matter, iron or manganese, and other trifling
impurities. We shall again notice similar varieties in other
places. (See Excursion XI., Art. 72.)

40. Near the pitchstone vein the high ridge of Dun-fion
reaches the coast ; and the trap which forms its summit
appears in section in the clifis, overlying the sandstone. It
presents a facade of imperfect columns, and contrasts strongly
with the stratified sandstone. Both sink rapidly southward
in the direction of the dip; the sandstone is depressed below

"8 GEOLOGY OF ARRAN.

the sea-level, and the trap then occupies the coast, forming
the low point at the entrance of Lamlash Bay, called Clach-
land Point, and also Hamilton Rock, the small island in the
channel. Near the point where the sandstone disappears,
some remarkable effects of trap dikes are exhibited. On the
west side of the point the sandstone suddenly emerges again,
and extends to Lamlash. The whole bay, indeed, has been
excavated in this red rock, which forms the base of the cone
of Holy Isle, as well as the coast on the mainland. But we
turn here meanwhile, and direct our steps to Brodick, by the
summit of the ridge which divides the two bays. Its
southern side forms a long grassy slope of gentle inclination
towards Lamlash Bay, on which are many granite blocks,
and a few of felspar porphyry ; towards the north it falls
suddenly in steep cliffs and terraces. As we pass up the
easy ascent striking views open southwards. Starting sud-
denly from the water's edge to the height of 1030 feet, Holy
Isle, with its encircling sea-line, fills the foreground grandly.
The bay, with its fine double sweep to King's Cross Point,
its wooded banks, debouching glens, and background of dark
hills, forms the right of the picture ; to the left the glassy
sea sleeps in the sunlight, dotted with small white moving
specks, and bounded by the winding line of the Ayrshire
coast, with promontory, creek, and bay, along which the eye
may range from Ardrossan heights to the Mull of Galloway.
The scene is singularly sweet and picturesque ; without
Holy Isle as an integrant part it would want character ; this
stamps upon it peculiar features. The northern mountains,
so essential and expressive in most Arran landscapes, are
here hidden from us ; but when we gain the highest edge of
the ridge, they burst upon us with startling suddenness in an
aspect quite new. They are grouped in a way not seen from
any other point, and their jagged profiles are thrown into
lines of singular boldness. Not less new and striking is the
aspect of the lovely bay of Brodick, with its noble castle and
hanging woods, of the glens into whose far depths the eye can

EXCURSION II. 79

reach, and of the " cottage homes " nestling amid groups of
trees, in shelter of the hills and sloping banks which enclose
the smiling fields of Brodick plain. Often as we have come
to the edge of the ridge by this route, we have always felt
the same delightful surprise when the scene first burst
upon us.

41. This ridge consists chiefly of sandstone. Trap rocks
form a thin capping along its highest part ; they extend a
little way down the southern slope, thickening as they
descend, but along the highest ridge are so thin that sand-
stone occupies in some places depressions in the ridge ; and
the trap occurs only on the isolated tabular knolls into which
the ridge is cut up, especially towards the west. The trap
consists of felspar and augite mainly, with imbedded zeolites.
MacCulloch designates it augite rock ; but hornblende and
iron also occur. The sandstone close to its junction with the
greenstone of the summit is highly metamorphic, resembling
a quartz rock. Along the steep northern front also, west of
Dun-Dhu, beds of clinkstone crop out in various places, at
different heights. These are either of truly igneous origin, or
are metamorphic sandstones, altered by whin dikes, of which
there are several, or by the near proximity of the great
igneous masses — greenstones, porphyries, and pitchstone —
which pervade the sandstone in this quarter. The association
of these various products affords one of the most interesting
sections to be met with in Arran. It is best exhibited a
little to the west of the path which leads down from Dun-fion
to the farm-houses. Dun-fion, or Fingal's Fort, is the high-
est point on this portion of the ridge, right over Corriegills
shore, and 500 to 600 feet in height. A low mound, enclos-
ing an elliptic space 40 yards by 16, is seen round the
summit, but nothing whatever is known of the history of the
fort. From traces of vitrification said to have been noticed
on the stones of the mound, it has been conjectured that the
place was the site of a beacon fire, in the wild old times, as
well as a stronghold. We were unable, however, to find the

80

GEOLOGY OF ARRAN.

least trace of such an effect ; but certainly the situation was
well chosen, if such was the purpose. It looked out far and
wide across the waters of the frith, so that no sail could pass
or approach unseen ; while the glare of its fires would light
up every glen and mountain-brow on this side the island,
from Dippen to North Sannox.* Dun-Dhu is a prominent
hill nearly as high, a little farther west, and standing out in
front of the cliffs, to whose base it is joined below. This hill
is composed of columnar felspar porphyry ; and between it
and the path descending from Dun-fion the beds represented
in the annexed diagram (fig. 17) occur in a well-marked ver-
tical section, — two veins of pitchstone and one of porphyry,
with sandstone intervening, surmounted by augitic trap.

Fig. 17.

(a a a a) Sandstone ; (6) felspar porphyry ; (c c) pitchstone ;
(d) overlying trap.

The upper pitchstone vein, c, is about thirteen feet thick,
and dips south-west, at a small angle, probably 10° or 15°, and
may therefore be conformable to the sandstone which is seen in
the quarry near the footpath to dip about south-west at 10°.
The rock is of a dark green colour, without felspar spots,
divided into beds from eight inches to three feet in thickness ;
and these beds consist of closely aggregated prisms, or laminae

* Headrick's Arran, pp. 64, 76, Statis. Acct.; Pennant (1772) does
not notice it.

EXCURSION II. 81

splitting into prisms. The bed is very high 011 the cliff,
coming close up to the prismatic greenstone, but apparently
separated everywhere by altered sandstone. Its upper sur-
face is on a level with some of the depressions in the edge of
the trap ridge; and it is apparently through the most eastern
of these that this bed passes across to the back of the ridge,
and is again seen behind Dun-fion, where a bed of sandstone
is also seen. It no doubt passes downwards towards the bay,
but it is not seen again : perhaps the trap of the cliffs is a
mere facing, and the pitchstone may traverse the sandstone
only. It is the upper surface of the bed that is here
exposed.

The lower pitchstone vein, c, is fourteen feet thick, and of
the same colour and structure as the other ; but its low level
can hardly admit its passing across the ridge. The vein of
porphyry occupies a ledge about fifty yards lower down :
this ledge being on the level of the front of Dun-Dhu, from
which the porphyry extends. The sandstone strata are not
well seen on the ledges a a a, between the beds, in the line
of section, but come out distinctly on their continuations
east and west.

Neither of these pitchstone beds has any connection with
those upon the shore, which are far below the level of the
base of our section. But the upper bed probably passes
westwards across the hollows south of Dun-Dhu, since a bed
of pitchstone, about twelve feet wide, again appears on the
west bank of the first burn north-west of this hill, and in a
position to which a slight undulation in the upper bed would
bring it. A few hundred yards farther west, in a field
within a high fence, it breaks out again with a breadth, of
more than twenty-five feet, and, undulating again, passes Tip
westwards by the crest of the moor, where it appears of
considerable breadth, there being only indications of its
occurrence between these two points. But from the crest
of the moor it can be traced continuously into Birk Glen.
(See Excursion IV.) The rock is most usually glassy, of a

82 GEOLOGY OF ARRAN.

bottle-green colour, homogeneous, and very seldom exhibits
any porphyritic structure. Between the two first outbreaks,
on the bank of a stream, there is a bed or dike of a very
hard and tough hornstone, or brownish gray quartzite, with
imbedded bits of quartz, a slight modification merely of the
pitchstone base; there is a greenstone dike beside it. The
relations of these igneous products are very curious.

42. The northern front of Dun-Dhu consists of fine
columns of porphyry, divided at irregular intervals by flat
joints. On the north-west side the columns are very perfect,
and have a singular diverging fan-shaped arrangement. We
ascend the hill easily from the south side, and find the
summit composed of huge prismatic masses of porphyry,
lying closely side by side almost in a horizontal position, and
in a direction from north-east to south-west, so that their
ends, cut uniformly off, form a wall towards the west, ten to
fifteen feet in height. Some of these columns are twenty
feet in length, apparently without joints : in general, how-
ever, there is an indication of a flat jointing, at distances of
four to seven feet. Most of the prisms are four and five-
sided. The rock has a gray felspar base, occasionally iron-
shot throughout, or merely streaked with iron; and the
imbedded crystals are of glassy felspar. Bits of quartz are
also disseminated through the base — apparently minute
crystals with their angles rounded off, as if by attrition.

No granite blocks were noticed on this detached summit;
but many of considerable size, as much as from six to four-
teen tons, were observed close to the cliffs on the west side.
They are found in great numbers, and of large size, over all
parts of the open plateau between the base of this ridge and
the shores of Brodick Bay. They have been mostly blasted
hereabouts, gathered off the fields, and reared into fences.
Many of them are rounded, polished, and marked with very
perfect glacial strise — traced upon them ere yet detached
from their native beds in the granite nucleus, or else in their
transit upon the ice to their present resting-place. Across

EXCURSION II. 83

this plateau our path homeward is by the tortuous lane
which enters on the Lamlash road immediately below the
Free Church. It pursues its winding way between fragrant
hedge- rows, through birky places, by burns, past cottages
with their corn-fields and meadows, and affords besides, in
its distant outlook, much to beguile the way. The evening
tints of rich purple-gray on the northern mountains, beyond
which the sun is now going down, and the profile of the
rugged peaks against the golden sky behind, are glorious to
look upon ; the light is strong yet on the peaks of Ben-Yim
and the Cobbler, on Benlomond and the mountains of Aber-
foyle; the broad shadow of Maoldon falls dark athwart the
sea ; but beyond this a flood of light comes streaming down
the northern channel, and brings out strongly the gray rocks
of Garrochhead and the Cumbrays, and the lovely glades
amid the woods of Fairlie. The turrets of Brodick Castle
are yet gilded by the sunlight which falls in broad sheets
across the lovely bay, whose glassy expanse is crowded with
fishing parties. Before we can reach our stately but pleasant
inn, the steamer has rounded Merkland Point, disappeared
iu the shadow of the hills, emerged again into the " lanes of
light," and threaded her way through the fleet of boats.
Already the fair throng has left the landing-place, scattered
now in gleeful groups along the various roads.

84

EXCURSION III.

TO THE SUMMIT OF GOATPELL.

43. GOATFELL is an unmeaning corruption of the native
name of this mountain. Gaoth (ga.5) is the Gaelic word for
wind; this may be the origin of the first part of the name;
with the animal indicated the mountain has no sort of con-
nection. Then, Fell is not a Scottish word; it belongs to
the North of England, and to Scandinavia in its form of
fjeld or field, .applied appropriately to the wide flat mountain
plateaux of South Norway. Bein, or with the aspirate
Bhein (ban or ben, ven), is a mountain. Pen is the English
form, as in Penyghent, Pendle Hill ; and hence the Latin
term Pennine for a principal range. Thus the name would
be Gaoth-bhein, or Bein-gaoth — the hill of the winds — not
very expressive or special as regards this hill more than
others standing prominent. But those who gave the name
perhaps knew no higher hill; and there is a peculiar effect
often seen here to which the name may perhaps have
reference. The Ben-Ghnuis or western ridge first arrests
the vapours ascending from the western sea, and condenses
them along its winding summit into a dark sinuous bank,
from whose shattered edge masses float away when the breeze
gets up, and dashing against the flat side of the ridge of
Goatfell, are driven in rapid eddies round its south end or
over its upper edge. To one looking up from the quiet
depths below, this would suggest the existence of a furious
gale upon the summit. As the weather thickens and the
clouds accumulate, the cone gets completely hidden, and the
rolling vapours pass even lower than the mill-dam, veiling
the edge of the great waterfall, which then seems to issue
directly from the clouds.

EXCURSION III. 85

But such a day as this will not suit for our walk; we must
wait for bright skies and still air — few walks in Arran will
then please us more. Every lover of mountains has a keen
desire to climb the highest summit within his reach, and
many will choose this walk for their first excursion.

The woods which stretch westwards from the castle are
crossed by two paths, by either of which we may pass upwards
into the moors. The readiest way is by the new approach
to the castle, which we enter by the gate near the bridge
over the Rosa burn. We keep on the avenue as far as the
west end of the stone bridge, then turn to the left through
an iron gate; and following the path a little past the
wooden bridge, again turn to the left, and soon emerge
upon the moor. The entrance to the other path, is by
the castle stables and old hotel, more than half-a-mile north-
east of the Rosa bum bridge. This meets the other path
before it opens on the moor. Or from the mouth of Glen
Rosa, reached as on our first excursion by Glen Shirag, we
can easily climb up the eastern flank of Glen Shant Rock.
A path lies near the east bank of the Cnocan burn, as far as
the mill-dam, where this stream is gathered from many
heads. Beyond this a track is marked out among the granite
blocks to the east shoulder of the mountain; thence, along
the edge of the ridge, a rugged path conducts us to the
summit, over huge masses of rock, and along the edge of
Cyclopean walls. But the geologist has much to see on the
ascent which this route will not shew him, and we must
conduct him by another way.

44. The strata which are exhibited in a longitudinal section
on this side the island northwards from Brodick Bay, are
shewn in the annexed figure (fig. 1 8). Here a a is the coarse
granite of the nucleus, extending from the mill-dam to Glen
Sannox ; b b, the slate penetrated by granite veins ; c c, old red
sandstone; d d d, carboniferous sandstone enclosing beds of
limestone e, seen at the church and in the woods N.W. of the
Castle. — The sandstone with limestone beds is succeeded by the

8G

GEOLOGY OF ARRAN.

'

Old Red as we ascend the burn above
the wood, the junction being somewhere
about the place where the wall enclosing
the new plantation abuts against the
bank of the stream. But the contact
is not seen, nor is there any gradation
visible. The Old Red is here a hard
quartzose slaty sandstone, with many
thin brown laminae, and elliptic blue
or white claygalls; the dip is back
against the slate of the mountain, con-
trary to that of the overlying carbon-
iferous beds, at angles varying from 55"
to 70°. As we advance the strata become
much obscured by debris; and a little
above the point where the west-burn
enters the main stream the dip is
reversed or towards S.S.E., at about
G5° to 70°. The rock is here coarser,
and contains imbedded masses of resin-
ous quartz and dark blue slate: farther
on it is darker and finer, with claygalls,
and the dip again appears to be towards
the slate, but the stratification is obscure.
In front of the first waterfall, amid a
group of birch trees, a mass of quartz-
iferous hornstone porphyry lies across
the bed of the stream, but seems to
terminate against the bank on the west
side ; on the east side its extension cannot
be traced, so that most probably it is
not a dike. It is a beautiful rock, and
if readily obtained in quantity might be
used for ornamental purposes. The base
is a dark reddish hornstone, containing
crystals of glassy felspar, and round

EXCURSION III. 87

bits of qtiartz. There are also beds of conglomerate ; and
the strata lean toward the slate, but less " end on ; " the dip
being about north 40° west.

We are now approaching the junction of the old red sand-
stone with the slate ; but as this is not well seen in the bed
of the main stream, it will be more instructive to diverge to
the left, up the course of the west branch, as far as the dark
brow of slate, where the hill suddenly rises. Passing up the
bed of this west burn, we find the common red rock and
conglomerate succeeded by flinty or quartzose sandstone,
obviously metamorpbic. The cause of this change is soon
discovered. At a waterfall on the burn there is an outburst
of a peculiar granite amid the sandstone strata. This is an
intimate mixture of quartz and felspar, without mica; in fact,
the Eurite or Weiss-stein of mineralogists. It extends for
many yards in the bed and banks — how far cannot be deter-
mined. Between it and the rocky brow which is formed of the
common dark slate, there are various metamorphic beds, con-
ducting us by insensible gradations into the true slate.
Some of these are white and gray flinty slates, others fine-
grained, hard sandstones. It is thus difficult to decide to
which series the beds ought to be referred. The strata have
been assimilated by the metamorphic action to which both
series have been alike subjected. A similar case occurs at
the junction on the White Water, to be noticed on another
Excursion.

Returning now to the bed of the main stream, and entering
it some way above the waterfall at which we turned off, we
find similar white and gray quartzites about fifty yards in
front of a deep chasm, with a waterfall, cut out of the slate
rock. The sandstone is not seen in close proximity; but the
junction must be near, as this is the first point at which the
slate appears in the bed or banks. Two dikes of greenstone,
from opposite sides of the pool, unite at the edge of the fall,
twice bifurcate, and twice unite again, enclosing two long
elliptic masses of altered slate, and then continue as one dike

GEOLOGY OF ARRAN.

right up the chasm. The branches are three to four feet wide ;
and the joint breadth, as one dike, eight or ten. The dike
undulates in its course from 30° to 40°, conforming to the
course of the chasm. The chasm is, in fact, due to the dike;
an original depression, produced by a fault or the irruption of
the dike, determined the channel for the stream, and along the
course of the dike the water met with least resistance in its
work of disintegration. The chasm is nearly half-a-mile in
length, with perpendicular walls ten to fifteen feet high,
above which the banks rise very steeply on both sides. It
runs in against the great sheets of slate, forming the waterfall
below the mill-dam. At the base of these sheets the dike is
seen again, interrupted or broken off in one place by the slate,
from beneath which it again emerges, and appears upon the
high brows above in the bed of a small stream entering from
the north-east.

45. We are now at the famous junction of granite and
slate close to the mill-dam. The appearances have been
often described. We shall quote the very clear account
given by Professor Ramsay: — "The absolute junction of
the two rocks is not here visible ; but that it is in the imme-
diate neighbourhood, probably in the bed of the dam, is
clearly shewn by the appearance of a granite vein, about one
foot broad, which penetrates the strata, and crosses the bed
of the stream about ten yards below the artificial wall which
confines the water of the dam ; thus indicating its intrusion,
while in a state of fusion, into the stratified deposit with
which it came in contact. The granite is of a yellowish
colour, fine-grained and compact in texture, and consists
principally of felspar. The slate is exceedingly tortuous ; and
the strata are intermingled with numerous veins of quartz of
varying sizes, and which generally alternate with the slaty
strata in regular minute laminae." This description is correct
and well stated; but an important fact has escaped notice
altogether. It has not, indeed, been alluded to by any one of
the many observers who have visited this locality, owing

EXCURSION III. 89

probably to the state of the water in the river at the time
when the junction was examined. We refer to a dike of
greenstone which crosses the river diagonally, ranging about
N. 20° E., and about thirty feet wide. It enters the east
bank of the stream under the mound or wall of the dam, and
is seen again on the surface, a little way toward the north-east,
but is soon lost under the heaps of granite blocks. The strata
of slate range N. 65° E., or almost east and west by the
compass, dipping S. 25° E., at an angle of 70°. Thus the
direction of the dike makes an angle of 40° with that of the
slate. Now, there are granite veins in this dike ; and these
cross out into the slate on the east side of the dike, continu-
ously without a change of direction. There is no mistaking
the rocks, both the greenstone arid slate are perfectly well
marked ; a portion of the former, traversed by the veins, has
distinct acicular crystals of white felspar, and that concentric
structure so peculiar to this rock. The slate is quite homo-
geneous— a dark blue hard quartzite, in some specimens
resembling Lydian stone — but the alteration from its ordinary
state is not remarkable. The principal granite vein is several
inches (six to eight) wide, and is traceable in the greenstone
through a space of fifteen or twenty yards, and runs right on
into the slate. Several smaller veins ramify through the
greenstone, dividing and then uniting again. These have the
structure so well described in the above extract. We con-
clude that the slate was first injected by the greenstone dike,
and that the irruption of the granite, at a subsequent period,
pierced through both of these rocks. The case is very
interesting, being the only one in Arran, that we know of,
in which greenstone is proved to have been erupted anterior
to the injection of the granite amid the strata of slate.

46. We recommend the geologist, who is not deterred by
the prospect of "a pretty stiff pull," to mount at once by the
southern shoulder of Goatfell, avoiding the common pathway,
already pointed out, which presents much less to interest
him. He will thus have an opportunity of examining those

90 GEOLOGY OF ARRAN.

huge natural ramparts of granite blocks, piled mass on mass
to a great height, like the Cyclopean walls of Tadmor or
Heliopolis, and of studying the structure of a granite moun-
tain, the varieties of the rock, and its peculiar style of de-
composition. We especially recommend this course if he
has not accompanied us on our first excursion.

We have now reached the south summit of Goatfell, the
highest point in the island, elevated 28G6 feet above half- tide
level. The north summit is 238 feet lower. There is not
perhaps in Scotland another mountain peak which looks
abroad upon a scene combining the same variety of grand
features. Many afford wider and finer mountain views ; here
the eye ranges over a vast extent of broken coast, the whole
expanse of the noble frith, and its many narrow branches
winding far in amid mountain solitudes. How grand from
this commanding height are the surrounding peaks and rugged
ridges, and the profound dividing glens ! what lovely pictures
in their glassy frames are these sister islands ! how stirring
the rapid movements of life all day on the inner frith ! what
a world of human interest in that great ocean steamer start-
ing on her outward voyage ! The " sweep of the circling
horizon" embraces a maguiticent amphitheatre, reaching from
the mountains of Donegal and Londonderry on the west and
south-west to Ben- Lomond and Ben-Ledi on the north-east;
from Ben-Nevis and the mountains of Mull on the north and
north-west, to the ranges of the South Highlands, the Mull
of Galloway and Isle of Man, in the opposite direction.

On a clear day, with a north or north-east wind, the
panorama of the northern mountains is very fine. On the
north-east horizon, Ben-Lomond is easily known by its ad-
vanced position, elongated form, and double top — west of it
on either side may be seen Ben-Voirlich (Loch Earn), Ben-
Ledi, and perhaps Ben-Lawers; north of it is the group of Loch
Voil and Loch Dochart, among which Ben-More is conspi-
cuous by its conical form. To the north-west of Ben-Lomond
is the Arrochar and Loch-an-Slui group, among which Ben-

EXCURSION III. 91

Voirlich, The Cobbler, andBen-Yim are the most conspicuous,
the last, farthest to the north-west, of an elongated form like
Ben-Lomond, but higher, and with two tops. Still farther
round to the north-west is the lofty group of Tyndrum, and
Breadalbane forest, among which, perhaps, Ben-Lui may be
recognized, with a deep corrie on its south-east side filled
with snow till far on in the summer. The wide extent of open
undulating countiy, without mountains, between Loch Fyne
and the sound of Kerrera, and bounded north by Loch Etive,
renders it easy to identify Ben-Cruachan on the east side,
with its double summit, and Ben-More in Mull on the west.
Ben-Nevis is less easily found; but having seen Goatfellfrom
it on a very favourable day, we are quite sure it can be seen
from Goatfell. It lies due north, far out on the horizon, and
a little to the west of Ben-Cruachan. The high group north-
east of Mull is that which lies between Corran Ferry and
Strontian ; north of this are three conical mountains belong-
ing to the district west of the " Great Glen of Scotland,"
somewhere near Loch Lochy. Carrying the eye southward,
we see the south promontories of Mull, with some low-lying
isles, perhaps Staffa or Coll, across the north end of Jm'a,
whose three paps are conspicuous and close at hand. South
of Jura is Islay, well seen across Cantire, whose two depend-
ent islands, Davar at the mouth of the Campbelton harbour,
and Sanda off the Mull, are under our feet. Across the top
of Ben-Ghnuis is seen Rathlin Island, south of which rises
the lofty ridge of Antrim, of which the most conspicuous
summits are Knoeklayde, Aura, and Trostan, nearly 2000
feet in height; in the centre of Antrim Slemish is remark-
able by its isolated position, and its form, a truncated cone.
The bold promontories of the north coast come out in succes-
sion towards the west, Bengore, Macgilligan, and Malin
Head; the top of the chain of Londonderry bounds the view
in that direction. Far out southwards the eye may sometimes
penetrate as far as the Isle of Man, and the groups of Morne
and Skiddaw. In that direction, twenty miles south of

92 GEOLOGY OF ARRAN.

where we stand, the grand cone of Ailsa rises abruptly from
the sea; beyond, the eye ranges far out south-east and south,
along the noble sweep of the Ayr and Wigton coasts. As
the day verges towards evening, a purple curtain slowly falls
over the scene to the south and east, while the light is still
strong on all the northern mountains. Towards the west
the islands stand grandly up in a sea of molten gold, which,
rising far out westwards, blends with masses of gorgeous
clouds "set on fire with redness" — later still, —

The sun, descending,
Leaves upon the level water
One long track and trail of splendour.

Of mere geological interest there is not much on the sum-
mit of Goatfell. Bits of greenstone lie about as if a vein
existed near. Veins of fine granite in the coarse variety, of
which the mountain consists, have been mistaken for veins
of claystone. These veins are clearly cotemporaneous, and
have no connection with the fine-grained granite of the
interior; it is too distant and too deep to have sent such
fine veins and threads to this height. Dr. MacCulloch
noticed a magnetic property in the granite, which he after-
wards confirmed by observations made on other mountains
in Scotland (West. Isles, ii. 351; Geol. Trans., 1st Series, vol.
ii. p. 430, and vol. iv. p. 124). The greatest deviation of the
needle which we noticed was on a block north of the cairn,
where it pointed to 75° west instead of 26° west, the present
variation ; a deviation of 49° from the true position. Such
being the irregularity, it may be well to stato that the tongue
of laud between Loch Fyne and the Kyles of Bute bears due
north from the summit of Goatfell. Ailsa Craig nearly one
point east of south.

47. If the day is calm and clear, one is disposed to linger
on the summit and take in the full beauties of the scene.
The Cyclopean walls on the north front are a fine study,
shewing the structure of a granite mountain, and the style

EXCURSION III.

93

in which the rock decomposes. About twenty courses of
huge slabs are piled up into a lofty mural precipice. A walk
is recommended down the north-west front to the dividing
ridge, past the north summit, to have a better view of the
corries, and a peep into the dark depths of Glen Sannox.

Fig. 19.
Cyclopean Walls.

The first col on our descent is marked by a dike of gray trap,
about twenty-five feet wide, with felspar crystals, giving it a
porphyritic structure, whose disintegration has determined
the formation of the col. It cuts the Ceims, on the opposite
side of Glen Rosa, and is there marked by a depression ; the
range is about west 25° south. Another dike ten feet wide,
with, the same range, is seen a little north of the north top
of Goatfell ; continued across the glen, its course is right up
the eastern front of Cior-Mhor. The wild grand cliffs along
here are formed of huge sheets, rising tier on tier towards the
west, and dipping gently back east in the direction of the
rims of the corries. We might easily descend by this route to
the Sannox road, either by the south base of Cioch-na-h'oighe,
or farther south by the hill-sides over the village of Come,
to meet the evening boat for Brodick. But we must return
back by the summit of Goatfell to examine the corries and
descend another way, more picturesque and full of instruction.

94 GEOLOGY OF ARRAN.

We have already attempted to account for the origin of the
glens (Art. 32). Denudation by water, acting along fractures
when the land was rising from beneath the sea, seems the
most probable cause. Some geologists would ascribe their
origin wholly to scooping out by ice; M. Necker, to their
being in the line of great dikes, along which, as in many
cases already cited by us, disintegration would be more rapid.
Whatever the cause, there must have been some original
difference of surface to determine the first action. The
origin of the corries is still more difficult to account for. It
has been supposed, that if such a bowl-shaped hollow existed
when the snow and ice began to become permanent on the
summits, ice would depend in great sheets from the rim, and
work all round on the rock to its disintegration and removal,
so as gradually to widen the corrie. But what gave the rim
in the first instance ? It must have been of considerable
size to bring into play this supposed ice action, much larger
than any hollow that could have been formed like those on
the Corriegills shore, already alluded to (Art. 34), as exca-
vated by the constant rolling about of large stones : so small
a depression could scarcely determine the beginning of ice
action. Volcanic action has been suggested but may be
dismissed as inadmissible, since we know that granite was
not erupted as volcanic matter. We can then only
conclude that the same cause which removed the mantling
slate from over the granite (Art. 13), and began to mould
the peaks, may have determined the first formation of the
hollows.

As regards the after work, we are inclined to adopt the
view very ably argued by the Rev. T. G. Bonney in a recent
paper on Alpine cirques, which are very similar to the corries
of our mountain tracts, that water, not ice, has been the agent
in their formation (Jour. Geol. Soc., Aug., 1871). Streamlets
from the upper heights trickling over the rim would "notch
the edge, seam the cliffs, and undermine the base with their
spray." In this way a number of small streams towards the

EXCURSION III. 95

sides and larger ones near the middle would form an amphi-
theatre rather than a straight wall. We have pointed out,
in another Excursion, the great power of excavation due to
the height of fall of even a small body of water.

The descent from the summit of Goatfell into Glen Eosa
and by the Rosa col must be carefully selected, as it is very
dangerous in some places. A few years ago a young man
lost his life by a fall on this descent ; and, more lately, a gen-
tleman from London, of literary celebrity, left behind by his
party, got into such a position that he could neither descend
nor mount, and had to remain in durance over night, till, dis-
covered by his answering shouts, he was released next morn-
ing by his friends who had gone up from Brodick to search for
him. — For the geologist, two descents are more instructive
than the rest. One of these is eastwards by Maoldon and
the White Water, but it implies rather more fatigue than
most persons will consider consistent with enjoyment ; and
we shall therefore reserve for another Excursion an account
of what is to be seen on this route. The other is by the
plateau south-west of the mill-dam, along the boundary of
the granite and slate, where some fine junctions, granite
veins, and contorted schist, may be studied. The line of
junction descends the north-west front of Glen Shant Rock
towards the mouth of the Garbh-Alt in Rosa burn, by which
we passed up the opposite hill in our first excursion. On
the edge of the plateau, a quarter of a mile south-west of the
"mill-dam, a good example of a " roche moutonnee " may be
seen. The exposed horizontal surface is fourteen or fifteen
yards in extent, rounded neatly off diagonally to the laminae ;
the north-west front, or vertical face, is also ground off, but
the opposite face turned south-east retains its original form.
This indicates the direction of the abrading force. The
whole surface of this plateau ought to be carefully searched
for glacial striation, and other examples of " roches mou-
tonnees."

96

EXCUESION IY.

THE WAYSIDE MUSEUM.

48. THE student of geology will make but little progress
in this delightful science without a thorough practical ac-
quaintance with the mineralogical distinctions of rocks. He
must first know a few of the common minerals — quartz and
its varieties, as rock-crystal, amethyst or purple quartz,
smoke quartz or cairngorm, hornstone, jasper and pitchstone ;
felspar and its varieties, as glassy and compact felspar and
albite ; mica, talc, chlorite, serpentine, hornblende, hyper-
sthene, augite, olivine, and calcspar. All these are "simple
minerals" to the mineralogist; the chemist calls them "com-
pound," since he can resolve them into elements. Using,
then, the terms simple and compound, in the mineralogical
sense, we divide the rocks into two classes; common slate,
limestone and the marbles, coal, quartz rock, and serpentine,
are the simple rocks; all others, forming by far the greatest
proportion of the mountain rocks of the globe, are composed
of two or more of the five simple minerals — quartz, felspar,
mica, hornblende, and augite. Knowing these, the student
will easily make out the composition of almost all the rocks,
certainly all those of Arran. With every term a definite
idea will then be connected, and the study of the subject will
be smooth and pleasant henceforward. Arran has been called
an epitome of the globe; our Wayside Museum is an epitome
of Arran, in so far as a collection of rocks is concerned. In
a few visits to this collection, the student will gain a know-
ledge of rocks, which he could never acquire from the hand-
specimens in a common collection — a knowledge absolutely
necessary to all true progress in the science.

49. The collection of rocks which we designate as the

EXCURSION IV. 97

Wayside Museum is easily visited. Leaving the village of
Lower Invercloy at the smithy, and passing up the hill along
the old Lamlash road, we are upon it at once. It is simply
a stone fence bounding the road on the east side, and reach-
ing from the brow of the old sea-cliff to the top of the hill.
The rocks of which it consists have been gathered from the
adjoining fields, across which they were borne from the
mountains and hills to the north and west, during the long
period when the island was wrapped in its icy sheet.

As we pass up from the level of the sea to the top of the
terrace— a height of five-and-twenty feet — the geological and
picturesque features of the mountain nucleus come grandly
out. In front the granite peak of Cior-Mhor and serrated
granite ridges of the Ceims (Kyims), Ben Tarsuin, and Ben
Ghnuis, tower above the flat- topped, steep-sided foreground
of slate and Old Red. On the extreme right, Maoldon, with
its rounded top of fossiliferous sandstone, rivals in height the
slate plateau south-west of Goatfell. Planed off across the
line of dip, and in the line of strike, these sandstone beds,
with their intercalated limestones, stretch along the uniform
line far down to Merkland Point, while their gentle south-
ward inclinations form the charming woods and glades about
the castle. Subsiding beneath the alluvial plain of Brodick,
whose bay, excavated in their yielding sti-ata, once reached
as far as Glen Rosa, these sandstones, bearing the limestones
with them, have been heaved up into a long narrow ridge,
which culminates on the extreme left in the rugged porphyry
of Windmill Hill and rounded top of Ploverfield, whose
granites and syenites reach thence across the moors to points
out of sight, by the heads of the Ben-leister and Clachan
Glens. As we pass to the summit level of the road, the
Sheans come in sight — knolls of diorite capping sandstone,
which isolates them.

50. As the position of the rocks in our Wayside Museum
is wholly fortuitous, we shall notice them in the order of
their age. Granite occurs in several varieties, exhibiting

98 GEOLOGY OF ARRAX.

distinctly the three ingredients, quartz, felspar, and mica.
The felspar is either white or red, more usually white in Arran,
being always in such quantity as to give its colour to the
rock, and is easily known by its laminated structure and
rhombic form, which presents, in a fractured specimen,
almost a square face. The quartz is harder, gray, or vitreous,
and, if in crystals, the form is a six-sided prism, terminated
by a six-sided pyramid; the sides of the prism being
finely striated across, those of the pyramid unstriated.
No specimens are met with containing the smoke quartz
or cairngorm crystals, which ate found on the back of
Suidhe-Fergus and the Castles, and are of the same
form as the common rock-ciystal, or " Arran diamond."
The mica is dark-gray, dark-brown, or almost black, in
thin, flat, four-sided prisms, bevelled at each end with
two surfaces; so that when these are equal in length to the
others, the crystal becomes a thin six-sided prism. When
hornblende takes the place of mica, and the rock consists of
quartz, felspar, and hornblende, it is called syenite. Horn-
blende is easily known from mica ; the crystal is a long
narrow prism, not divisible, like mica, by the point of a pen-
knife, into thin light flakes, and the colour is darker. Syenite
has a mottled, dark-green, red, or white aspect, accord-
ing to the colour of the felspar, and is much tougher than
common granite. We have not met with it among the
northern mountains, and it may have found its way here
from the Ploverfield district. Some syenites contain also a
little mica; but such structure has not been noticed by us
in Arran. Another variety found here consists of quartz
and felspar, and has either no mica or extremely little; it is
called Eurite, and is usually tougher than other kinds of
granite. It is the Weiss-stein of the Germans, and the
Elvan of Cornish miners.

We do not find gneiss in our Wayside Collection — it does
not exist in Arran, and is only found in situ far to the north-
west. Mica slate, consisting of quartz and mica, which occurs

EXCURSION IV. 99

on the north-west coast, and in Cantire, is met with, as are
the common slates enveloping the granite nucleus, among
which the most remarkable is a very hard and tough banded
slate, such as occurs in situ near the line of junction of the
granite and slate, in the various localities already indicated.
The bands are no doubt due to a re-arrangement of the
colouring matter, probably iron, caused by the action of heat
along the line of contact. Both parts are equally flinty.
Masses of slate also occur composed of quartz and chlorite,
chlorite taking the place of mica. This chlorite is of a green
colour, and occurs in small very thin flakes, never reaching
the size of plates, as mica does. There is very little of this
chlorite slate in Arran ; but it occurs in North Bute, alter-
nating with mica slate, chiefly towards its outer border. A
few masses of another slate occur, which consists chiefly of
hornblende, but dashed occasionally with spots of felspar; it
is almost like a basalt in structure, but coarser in grain,
and more crystalline. It is extremely hard and tough.
This is hornblende rock ; it probably forms beds in the slate
on the north-west coast, but we know no locality for it.
Masses of red sandstone and conglomerate are frequent.
The chocolate-coloured Old Red, such as is well seen towards
the entrance of Glen Sannox, occurs, and another singular
variety, a passage apparently from slate to sandstone. A
mass of this will resemble porphyry in one part, in another
it will contain pieces of imbedded slate, and again there will
be finer layers, like a fine semi-crystalline conglomerate or
coarse slate. Large lumps of quartz are abundant, of several
varieties; some derived from the conglomerates, and others
directly from the northern mountains.

51. By far the most interesting series of specimens to be
found here is that embracing the igneous rocks of later origin.
These rocks affect a stair -like arrangement of beds when
forming the surface of a district, sheet succeeding sheet
upon the ascent of a hill, in a terraced form, like the steps of
a stair. They are hence called Trap rocks, from the Swedish

100 GEOLOGY OF ARRAN.

trappa, a stair. They are also called Plutonic rocks, as
formed in the " nether depths," to distinguish them from the
volcanic rocks, erupted at the surface.

The name Diorite is now usually applied to the rock
formerly called common greenstone, which consists of horn-
blende, and felspar usually of the variety called oligoclase;
Diallage rock and Hypersthene rock are the varieties in which
Diallage and Hypersthene take the place of hornblende.
Basalt is a nearly homogeneous compound of augite and fel-
spar, of a black or dark-blue colour, and great specific
gravity, owing mainly to the presence of 14 percent, of the
oxides of iron and manganese, and has several varieties
denoting merely difference of texture. Thus a coarsely
crystallized variety, in which the ingredients are distinct, is
called Dolerite, as it has then a deceitful resemblance to
diorite ; but if close grained, though still distinctly and
uniformly crystalline, is called Anamesite or intermediate;
basalt proper is compact and semi-vitreous, and has a perfect
conchoidal fracture. The composition of the fine varieties is
known, independently of analysis, mainly by the gradual
passage of coarser kinds into them. Ambiguity is best
avoided in naming these different varieties, by stating the
mineralogical composition and the grain. "We fail to see
the advantage, so far at least as British geology is concerned,
of the new names as above explained. Common greenstone,
the prevailing variety of those traps which consist of felspar
and hornblende, is quite as expressive as diorite ; and in-
stead of dolerite we have a name quite definite in the term
coarse basalt. But in truth the varying constitution of the
rocks defies all powers of nomenclature. Thus common green-
stone or diorite sometimes contains both augite and olivine,
as well as hornblende ; and in basalt we sometimes find
hornblende and olivine. But usually hornblende and augite
do not occur together; and augite may be regarded as a
fused and re-cooled hornblende, since the latter, when fused,
cannot be re-formed, but takes the form of augite, and since

EXCURSION IV. 101

augite (with some kind of felspar) is the invariable in-
gredient in recent volcanic rocks. Hornblende usually
shews a slight play of two colours, augite has one uniform
aspect. Amphibole is a general name for hornblende and
its varieties, as actinolite, tremolite &c. ; Pyroxene for
angite and its varieties, as diallage, hypersthene, and
sahlite. It is important to bear in mind, when study-
ing these rocks in the field, that however different they may
appear, they are all very similar in composition, the differ-
ences among them being due to different rates of cooling from
a state of igneous fusion. Rapid cooling produces a glassy
substance like the obsidian or volcanic glass of modern
volcanic districts ; but if the fused matter cool slowly it
crystallizes, a complete physical change takes place, and a
stony structure is assumed.

The chemical composition also of the greenstones and
basalts is very much alike, ranging from 45 to 60 per cent, of
silica, 10 to 20 per cent, of alumina, 10 to 15 and sometimes
25 per cent, of oxide of iron, with small quantities of potash,
soda, lime, and magnesia. It is chiefly the black basalt of Staffa
and the Giant's Causeway that contains so large a proportion
of iron. Other varieties of the hornblendic traps of Arran
are those improperly called trap-porphyries. They are fine
greenstones, containing imbedded crystals of felspar; and the
base being thus compound, they are not properly porphyries
One is speckled with felspar dots ; the other more homo-
geneous and dark-based, with distinct felspar crystals.

Diorite, or common greenstone, is also found in spheroidal
masses, which are very interesting, as illustrating the origin of
these rocks. The spheroids decompose in concentric coats, and
are found to be finer-grained inwards, so that the nucleus is
often very fine in the grain. As the crust thickened by
cooling, the heat would part from the centre more and more
slowly, and the most perfect structure would be developed at
the centre. This structure is indeed singularly characteristic
of the trap rocks, the slightest development of it enabling us

102 GEOLOGY OF ARRAN.

at once to recognize them amid inetamorphic slates ; the
columnar and jointed forms which these rocks so often put
on being also dependent upon it. All our basalts, green-
stones, and felstones exhibit this columnar and articulated
structure; but the concave and convex jointing is seen only
in basaltic pillars. It is plain that dikes are prismatic
across, because the heat passed off perpendicularly to the
cooling surfaces.

52. There are several varieties of felspar distinguished by
their forms and other characters, as well as by their chemical
composition. The common dull white felspar of the coarse
granite is orthoclase and a potash-felspar; a much whiter
kind, with smaller crystal and a jagged termination is albite,
a soda-felspar ; another variety is oligoclase or andesite,
a soda-lime felspar ; labradorite (with fine play of colours),
found in Skye, a lime-soda felspar, the properties being
reversed'; and lastly, anorthite or lime-felspar. Glassy
felspar in large clear crystals is usually present in the Arran
porphyries, as those of Drumadoon and the south-west coast.
The first three varieties only are those usually associated with
quartz in rock masses.

The felspathic rocks — those containing felspar alone — are,
viewed chemically, silicates of alumina, mixed with silicates
of potash or soda, lime, iron, and manganese; when horn-
blende or augite are in combination with the felspar, they are
silicates of magnesia, combined with the other silicates as
above.

The felstones of Arran are, felspar porphyry, claystone, and
clinkstone. When a rock consists of any simple base with
crystals of felspar imbedded, it is called a porphyry, and is
named after the nature of the1 base. Thus felspar crystals
being imbedded in a uniform felspar base, the rock is felspar
porphyry; so also we have pitchstone porphyry, hornstone
porphyry, and so on. But the term is often used loosely,
as when the base is not simple, and the imbedded crystals not
felspar. Compact felspar, claystone, and clinkstone are mere

EXCURSION IV. 103

varieties of one substance — a homogeneous felspathic rock,
with occasionally bits of imbedded quartz, and consisting of
about 75 per cent, of silica, 15 of alumina, and 10 of potash,
soda, magnesia, arid oxides of iron and manganese, with a
trace of lime. All these felspathic traps or felstones abound
in Arran.

Pitchstone, which figures so largely in the geology of Arran,
is found in several varieties in our " Museum ; " and when,
containing specks of white or red felspar, forms pitchstone
porphyry. Its composition is almost the same as that already
given for felstone, being from 62 to 74 per cent, of silica, 11
to 17 of alumina, with small quantities of potash, soda, and
lime, and oxides of iron and manganese. Its peculiar
character is derived from its rapid cooling. We have to
mention, lastly, a highly interesting rock found here, but
never met with by us in situ in Arran ; only a few blocks of
it are seen in the fence. It is intermediate between basalt
and pitchstone, and links them together as products of fire
generated at different stages of the cooling process. It is
hard and tough, and almost homogeneous like basalt, but
without its shining points, while it has the vitreous aspect
and colour of a dark-blue pitchstone. It would be extremely
interesting to find this rock in situ, and to mark its rela-
tions, whether occurring alone, with basalt, with pitchstone,
or whether all three are together. Veins of it must exist
somewhere to the N. or N.W.

53. Our walk homeward will be varied by passing down
from the top of the hill into Birk Glen, along the old road.
Having crossed the burn, we come on a great pitchstone
vein, lying a little diagonally to the path. It is of a green
colour, lamellar structure, and is in parts porphyritic. The
breadth is about thirty feet, and range between west and
south-west. Thin white films form on the exposed surfaces,
and when constantly moist decompose into a tenacious clay,
coating the surface. It extends into the bed of the burn, and
is there of the same width ; but the contact with the sand-

104 GEOLOGY OF ARRAN.

stone is not well seen. So far as can be judged there is little
alteration on the sandstone. On the lower side a dike of
hard, blue, fine greenstone is in contact with the vein ; but
the portion touching the dike is shivery and crumbling.
Passing upwards to the new Lamlash road, indications of the
vein are seen on the north of a claystone bed, which is
quarried, on the west side of the road. On the east side,
some way above the road, the bed of claystone, formed below
of rhombic blocks, but slaty above, is worked extensively for
road metal ; and on the north side of this quarry the pitch-
stone vein is seen coming against the claystone, and partly
covered by it ; a wing, as it were, of the claystone bed
stretches partly over the pitchstone, so that it does not
reach the surface — a curious relation rarely to be seen
among the Arran dikes. It is quite possible, therefore,
that it may be covered wholly or in part by the claystone
bed or vein below the road, since it is exactly in its line of
bearing. This relation of the two beds lends countenance to
the idea that these claystones are but altered sandstones. It
is remarkable that there is no trace whatever in the river bed
of this claystone vein. The pitchstone is traceable all along
the hill -side towards the N.E. to the crest of the moor;
whence it bends round nearly S.E., and reaches the base of
Dun-Dhu, as already pointed out. This course implies un-
dulations somewhat greater than any we have yet seen in
dikes, some of them being fully 30° ; but on the Corrie shore
they reach 80° ; and it is therefore quite conceivable, since
the continuity is made out so far, that our pitchstone vein
may bend round again after leaving Birk Glen, and re-appear
at Brodick school-house. Still it would be satisfactory to
find it at some intermediate point.

This pretty Birk Glen would be fit for cultivation ; but it
has been long kept in its present state of wildness as an
excellent cover for woodcock. We may return from it by
the new road ; but there is more variety in descending the
stream and passing through the ornamental grounds connected

EXCURSION IV. 105

with the hotel. The sandstone strata are well exposed, and
there are some curious dikes. The burn throughout its
whole course is highly picturesqiie ; indeed, none of the
Arran burns, lovely as most of them are, have the same
variety of picturesque features as those in the red sandstone.
The alternation of hard and soft strata, of sandstones and
conglomerates with variegated marls, the depth and peculiar
style of disintegration of the rocks, the many pretty ledgy
falls, the contrasts of the black basaltic veins, of the rich
colours of the rocks, the luxuriance of the wild wood and
wild flowers, bring out a variety in the effects not to be seen
in burns running over granite, slate, or basalt. This is
strikingly seen in the burns of North and South Sannox
glens, Glen Scorodale and Torlin, which gain greatly in
picturesque beauty directly on their entrance among the
sandstone strata. In drawing and painting them, rocks
ought always to be individualized ; and every artist ought
to be a geologist. A thorough acquaintance with the rocks
would be like the implanting of a new sense.

106

EXCURSION V.

BY THE CORRIE SHORE.

54. THE hotel at Invercloy occupies a charming situation
on a terrace of the old sea-beach. The site was chosen by the
late Duke of Hamilton, who had a true and exquisite taste in
art and nature, and the keenest relish for their beauties. Per-
haps the finest view is had on turning the north-east angle of
the building, where the granite peaks and ridges, the wooded
hill-sides, the castle, the sweep of the bay, and the opening of
Glen Eosa are all in sight. When the mountains are clear of
mist, the effects are always wonderful. They are perfectly
magical sometimes in the early summer morning, when a
gentle air comes into the bay from the cooler water outside,
and masses of white vapour cover the middle region of the
hills, the peaks projecting,

" Where many a pinnacle with shifting glance,
Through the gray mist thrusts up its shattered lance,"

and the lower edge of the vaporous mantle is like a veil of
gauze, half concealing the castle and its woods. Slowly the
veil rises, the vapour rolls upwards, now covering and now
revealing the highest peaks, till, as the bare rocks get heated,
the mist gradually melts away from the lesser heights, and all
that now remains is gathered on the sheltered side of Goat-
fell, and raised by the eddies from the glens and corries into
a cloudy pillar resting on the summit. Presently dissolved
below, the pillar in its upper part mingles with the fleecy
clouds aloft, and the peaks stand out all day long in un-
clouded grandeur.

From the hotel at Invercloy to the village of Corrie is a
distance of six miles ; but the sweep of the road round the

EXCURSION V. 107

bay takes up nearly two of these. A boat may be taken to
the Brodick pier opposite, the Rosa burn may be crossed by
a rustic bridge, but the sands will prove not a short route,
and the river is usually impassable. We shall pass round by
the public road. The Cloy burn formerly entered the sea at
the lower village, as the name Invercloy implies ; that course
was changed by a bar of sand cast up during a storm, and the
stream has now for a long time followed the winding course
which unites it with the Rosa burn before reaching the sea.
The ridge of gravel and sand dividing the beach from the
marsh behind is the joint work of the sea and the two
streams. This marsh, and parts of the gravel ridge which
nourish coarse brushwood, are the favourite resort of many
plants — as the Isle of Man cabbage, the sea purslane, the sea
pink, catchfly or sea campion, the scurvy grass, arrow grass,
two plantains, and many others.

Brodick church, on the platform finely overlooking the
plain, was formerly a chapel-of-ease to Lamlash church ;
but Brodick is now erected into a separate parish, in so far
at least as quoad sacra, purposes are concerned. For secular
purposes, the old division into two parishes still subsists.
That division was made in a singular way : the western half
of the island, from Loch Ranza to Largiebeg, formed Kilmorie
parish, the eastern half Kilbride parish, the boundary line
running along the crest of the dividing ridge of the island.
The north part of Kilbride parish forms the new parish of
Brodick, which extends northwards from the top of the
ridge between the bays of Lamlash and Brodick. This
division was made in 1864, and an endowment provided,
jointly by the late Duke of Hamilton and the Endowment
Committee of the Established Church.

Passing eastwards from the Rosa bridge, and then north
towards Brodick pier, and emerging from the belt of plant-
ing, we are on the site of the old village of Brodick, and
obtain one of the sweetest views in the island. The lower
reach of Glen Rosa, with its grand background of mountains,

108 GEOLOGY OF ARRAN.

is seen across the rich glades of the park, covered with
pasturing sheep, and finely varied with tree clumps and
tall standing stones. The enlargement of the park, the
formation of a new approach, and the opening up of this
fine view, rendered it necessary to demolish the village.
Yet had the old place a beauty of its own. The situation
was delightful ; and it was pleasant to see, on a fine summer
evening, as one returned from a walk among the hills, the
happy village groups enjoying themselves out of doors; the
young sporting on the common, or strolling in the green
lanes; the matrons bustling to and fro; the men in knots
about the cottage doors. But the household gods have
sought a new sanctuary at Upper Invercloy, and the popula-
tion is now out of sight.

55. Between the fine picturesque group of ash trees near
the beach and the Brodick pier, the strata of sandstone rise
from beneath the sands, and exhibit some extraordinary
irregularities of stratification. Two set of beds dip out of
the usual direction and at higher angles. The usual dip is
due south by the compass — that is, S. 26° E.; the irregular beds
make angles with these. Those on the east, next the regular
strata, dip E. 34° S. The other set west of these dip due
east; and when in the next set the usual south dip is
resumed it is 10° farther round toward the east — that is,
S. 36° E., instead of 26°. There is no apparent ca"use, in
dike or fracture, for this strange tumble-about among the
beds — it seems more probably due to eddies at the time of
deposition.

East of the pier, the strata are also slightly irregular, and
the angle of inclination greater than usual. The beds near
the pier dip S. 46° E., and a little farther east the dip is
S. 21° E. ; the former being a difference of 20° from the dip
west of the pier. We shall find similar examples of these
irregularities farther along the shore. By the roadside here,
under the summer-house perched aloft upon the cliff, there
is a huge granite boulder, the weight of which cannot be

EXCURSION V. 109

less than from six to eight tons. Somewhere about here,
or near the old hotel, past which is the most approved path
to Goatfell, there formerly projected over the cliff a very
large ancient yew-tree, called the "hanging yew;" hanging,
alas ! in a double sense ; for upon it, in the wild rule of some
of the old bailiffs or stewards, who in the absence of their
lords held despotic sway in Arran, summary justice, without
trial, was wont to be executed. But we must pass up to
examine the castle.

56. Brodick Castle stands on a terraced platform of the
sloping hill-side, at an elevation of 125 feet above the sea
level, towards which several successive terraces descend to a
distance of 400 yards, ending in the old sea cliff, here about
25 feet in height. The situation is thus fine and command-
ing, and the views from the principal apartments extensive
and beautiful. The building is in the old baronial style,
with battlemented roof, and is three storeys in height. A
lofty tower, with terraced gables, and flanked with turrets,
forms the south-west corner of the building. From the
battlements of this tower the sea and mountain view is very
grand. The alterations and additions made by the late
Duke, when Marquis of Douglas, and which were completed
in 1844, have fully doubled the capacity of the building,
and harmonize admirably with the ancient portion. The
entrance is by the west front ; and the building terminates
eastward in a ruined tower, clad with ivy and stunted
fuchsias, whose erection dates back at least 300 years, and
which is carefully kept in preservation. The handsome
grounds about the castle are laid out in a style admirably
suited to the situation. Off the south-east angle of the
castle there is a beautiful flower garden, sloping southwards
from the base of the first terrace, whose summit is crowned
with a row of splendid old plane trees; and there are many
fine limes, beeches, and other trees, in the older portion
of the woods about the castle. The disposition of the woods
towards the north has been greatly improved within a few
years by transplanting.

110 GEOLOGY OF ARRAN.

A stronghold of some kind seems to have existed from
very early times on or about this spot. It was a place of
great strength and considerable size in the time of Bruce,
who, after his descent upon Arran, from Rathlin, besieged
and took the castle, then held under Edward I. by Sir John
Hastings; and from Brodick he is said to have taken his
departure for the mainland, on his perilous expedition for
the liberation of Scotland. This was in 1307, and there
were still seven years before him of peril and varied fortune
ere his authority was finally established at Bannockburn,
on 24th June, 1314. The history of the various occupants
of Brodick Castle will be found detailed with consider-
able fulness in the Statistical Account, and Reid's
History of tfie County of Bute* The castle was razed
to its foundations, in 1544, by the Earl of Lennox, sent
by Henry VIII. on an expedition against Scotland — so
that no part of the present building can be older than that
date.

The connection of the Hamilton family with Arran dates
from 1474. Sir James Hamilton, created Lord Hamilton,
his manor house of Orchard, in the barony of Cadzow, being
declared his principal messuage, and having its name changed
to Hamilton, was attached to the Princess Mary, daughter
of James II., and would have espoused her but for the
untimely death of her father. Her hand was afterwards
bestowed by her brother, James III., on Thomas, son of
Lord Regent Boyd, and, as her dowry, a grant of lands in
Arran was made, and Boyd was created Earl of Arran.
That family, however, soon fell — their estates were forfeited,
the Earl fled abroad, and died there. In 1474, Lord
Hamilton, now a widower, married his first love, the
widowed Princess Mary, but survived only five years, dying
in 1479, and leaving a son, James, second Lord Hamilton,
then only four years old. By charter, dated 10th October,

* Messrs. Murray & Son, Glasgow,

EXCURSION V. Ill

1482, the forfeited lands of Regent Boyd and his son, the
Princess' first husband, were made over to her in liferent;
and in 1503, her son, now grown to man's estate, and having
already rendered important services to the Crown, received
(August 11) a charter *of the lands in Arran, and the title
of Earl of Arran; and by charter, dated the next day, a
gift " of all the castles and fortalices, mills and fishings, with
patronage of all churches, and a commission of justiciary."*
The honours and properties of the family steadily increased
from this time, fortune never for a long time forsaking
them. They never lost hold of Arran, the whole island
belonging to them, except the estate of the Fullarton family.
For about fifty years, Lord Rossinore held all the north-west
portion of the island, from lorsa water-foot to Loch Ranza,
except Whitefarland, part of the Fullarton property, having
received it, being unentailed, as a marriage dowry with
Anne Douglas, a daughter of Douglas, eighth duke. But
Lady Rossinore dying without issue, in 1844, the Hamilton
family soon after acquired the property. Anne, eldest
daughter of James, first duke, duchess in her own right
as heir of her uncle William, second duke, who fell at the
battle of Worcester, fighting on the royal side, did much
to improve Arran; and her benefactions are still remembered
by the islanders. She died in 1716, at the age of 80.
But perhaps no member of this family has done so much
towards the physical and moral improvement of the people
as the late duke, William Alexander Anthony Archibald.
He greatly improved the means of communication between
the different parts of the island, encouraged and aided
the farmers in draining not only the arable land, but also
the hill pastures ; built new hotels, built and endowed
schools, and endowed a new church. The improvement
of the castle and park are also due to his taste, and he
built the fine lodge on Dippen cliffs. His Grace's munifi-

* Reid's History of the County of Bute, p. 176 et seq.

112 GEOLOGY OF ARRAN.

cent patronage of art and exquisite taste are well known.
His Grace, when yet in middle life, died at Paris, July
15, 1863. His memory will long be cherished, as is that of
the Duchess Anne, by the grateful islanders.

57. Between Brodick pier and the gate of the principal
approach to the castle there are many curious dikes and
irregularities in the strata; indeed, the whole coast from
this point to Corrie is worthy of careful study, as illustrating
the natural history of dikes; and will prepare the geologist
to expect a broad dike to become a thread, and not to
be surprised if he see a dike "doubling" at an angle of
90° and " throwing" him completely " out." Between the
granite boulder and the gate there is one dike which is,
as usual, prismatic across, and breaking in westwards among
the strata becomes a bed parallel to them, and presents
on its upper surface a miniature "causeway" or pavement
of prisms of various forms, placed perpendicularly to the
surfaces of the bed. The case is very interesting, and
is a good illustration of the remarks made in Art. 52. A
little way east of this there are some other dikes on the
flat beach which change their directions in the most sudden
and puzzling way; and under the wall opposite the gate
a singular contrariety of dip in the beds; strata dipping W.
come "end on" against those dipping S. The dike here
which ranges due N. and S. by compass, and the others
already mentioned, may have something to do with this
irregularity. Near this point is the landing-place called
the wine-port, where in the jolly old times, when the
retainers were feasted in the castle, wine and beer were
brought ashore in hogsheads; and farther up, near the cliff,
are shewn the remains of an old burying-place, where some
important personages are said to have been interred. But
of this there is no special record.

It would require a separate memoir on this coast to
detail all the curious particulars that might be mentioned.
We can only state in general that in the bay here are

EXCURSION V. 113

some broad dikes with remarkable bends, greatly altering
the sandstone, and from their width and direction clearly
identifiable on the Strathwillan shore opposite ; that several
singular inversions of dip occur in the sandstone beds, whose
surfaces also are worn in, a singularly uneven manner, so
that portions stand out high above the others; that about
Carlo cottage and the smithy, and between this point and
the sandstone quarry, near Corrie, there will be seen several
dikes which bifurcate, and in the angle of separation enclose
altered masses of sandstone; and that equally striking cases
will be met with of a change in direction among these
dikes. After examining them, no one will hesitate to consider
it quite possible that the pitchstone vein of Birk Glen,
traced all the way across the moor from the base of Dun Dhu,
may be the same which shews again at Brodick school-house.

The old sea cliff is worn everywhere into caves by the
long-continued action of the waves. One of the largest
of these is " Lily's Cave," west from Carlo cottage, named
from a poor woman, of this name, who took it for her
permanent abode when her cottage was taken down to
make way for some former improvement, the date of which
is uncertain. She was a widow, with one daughter. The
youthful and pretty Lily had a lover in good circumstances,
whose truth stood the test of misfortune; but though he
had good hopes of Lily's favour before, she would not
listen to his proposals now — the widow would not leave
her retreat, and Lily would not desert her. But soon
the fair flower drooped and withered; and before a year
had passed, the widow was left alone in her new dwelling.
A gang of gipsies occupied the cave after the poor widow,
and became a trouble to the neighbourhood. They were
expelled by force ; and to prevent any new occupancy,
the gravel bank above the rock was digged down and the
mouth obstructed.

It is rarely that shells can now be found in any of the
caves or hollows of this cliff. The rapid growth of vegeta-

114 GEOLOGY OF ARRAN.

tion under the shelter, sunshine and moisture has covered
the floors with a deep deposit, in which it is hopeless
to search. We have added nothing to the list originally
given by Mr Ramsay, and which comprises the following: —

Cardium laevigatum. Venerirupis decussata.

Patella vulgaris. palaestra.

Lucina radula. Mytilus edulis.

Purpura lapillus. Buccinum undatum.

Turbo littoreus. Venus fasciata.

Turbo rudis. Cytherea exoleta.

Trochus cinerariue. Terebra reticulata.

magus. Rissoa calathisca.

crassus. seuricostata.

Nerita littoralis.

Three of these — the Venerirupis decussata, Rissoa calathisca,
and Trochus crassus — are now rare in the West of Scotland.
These species therefore seem to connect the beds in question
with those containing species now extinct in the seas around
Britain. The young geologist needs to be cautioned about
drawing any conclusion from a few scattered shells found on
the hills, as birds often carry shells even of considerable size
up among the mountains. It is only a regularly formed bed
that affords evidence of physical changes in the surface.

58. As we approach the village of Corrie, the sea cliff
becomes lower, and on the edge of it there rest two huge
granite boulders, one on each side the White Water, which
here comes tumbling down from the corrie on the north-east
side of Goatfell. They are true boulders, granite resting on
sandstone. The larger of the two, and the largest boulder
in the island, lies on the north side of the stream. We
estimate the weight at more than 2000 tons ! What force
could have hurled a mass so enormous from the mountain-
side? Not gravity urging its descent; unless, perhaps, the
first impulse was given by an earthquake, for the slope of
the hill-side is gradual, and the mass itself ill adapted for

EXCURSION V.

115

easy rolling. But if we admit the agency of ice, we have a
force adequate to the transport. This transport took place
before the present level of land and water was established ;
the cliff was not yet cut out from the sloping hill-side; the
bed on which the boulder now reposes was the sea bottom,
and in the glacial period it may have glided down upon icy
sheets descending from the conies, or been borne off by a
berg which deposited here the heaviest portion of its load.
A sister block rests upon the same cliff, on the south side of
the burn ; but its dimensions are much less. The shore is
strewed with multitudes of similar masses of all sizes.

From this point the geologist ought to pass up to examine
the limestone beds, thence to the top of Maoldon. and so
across to the granite junction at the fall of the White Water.
The limestone bands come out in the broken ground in front
of the north cliffs of Maoldon, dipping south-east with the
sandstone at angles of 30° to 40°. They are similar to the
beds at Corrie, with the same number of integrant strata,
each of which, however, is thinner than at Corrie, and with
the same assemblage of fossils. These limestone bands, as
well as those in the Brodick woods, have been supposed to be

d e

Fig. 20.

a, Granite of Goatfell ; b, slate ; c, old red sandstone ; d d, sand-
stone with coal plants ; e, limestone band ; /, tideway ; g, terrace and
old sea cliff; h, top of Maoldon, 1206 feet high.

upthrows of the Corrie beds, due to four great faults ; but we
doubt the correctness of this view for reasons already stated
(Art. 13).

If such a dislocation as is supposed really took place, there

116 GEOLOGY OF ARRAN.

ought to be evidences of it on the shore, where all the strata
are well exposed. Certainly none of the limestone beds of
Maoldon are seen upon the shore. We cannot agree with
the view stated by Murchison and Sedgewick, that there is
a series of advances of granite from the central ridge. From
the north side of Glen Sannox to the mill-dam, the line of
junction of the granite and slate is extremely uniform, un-
broken by spurs advancing from the granite nucleus. We
believe that when the paper on Arran by these distinguished
geologists was written, and even six years later, when Mr.
Ramsay wrote, the fact was not properly recognized, that in
the Scottish carboniferous system bands of limestone may
occur anywhere — from top to bottom of the series.

There is little else worthy of notice here; it is a wild
bosky place, encumbered with huge fallen masses of sand-
stone, and with granite blocks, among which many pretty
ferns and flowering plants find suitable habitats. But there
ai-e adders also about, and it is well to be watchful. The
north and north-east fronts of Maoldon are very precipitous ;
the western part of the precipice is intersected by a green-
stone dike about fifteen feet broad, ranging north 25° west
and inclined towards the east at an angle of about 15°. It
forms a deep chasm in the cliff, over which the sandstone
rises in a lofty wall. The amount of wearing here is pro-
digious ; and we cannot conceive how it can have been
effected without the action of the sea. Yet the place is
many hundred feet higher than the ancient level, indicated
by the raised beach so often mentioned already. We must
therefore call in the agency of the sea during the elevation
of the land at an earlier period. Through a narrow passage
between the dike and the sandstone wall, we can ascend to
the top of the cliffs. A remarkable alteration is produced
in the sandstone by this great dike. The top of Maoldon,
about 1206 feet high, is composed of soft red sandstone,
dipping S. 25° E. at angles of 15° to 20°; and thus the
strata are "end on" to the granite. Impressions of plants

EXCURSION V. 117

may be found in the sandstone. The summit is strewn with
granite blocks of moderate size.

In crossing from Maoldon to the fall of the White Water,
we pass the old red sandstone and slate bands; but no
junctions ai*e visible till we reach the base of the fall where
the burn issues from the wild corrie. Judging from the
position of the granite masses on the hill-side, one would not
expect the junction so far up — it takes place at the base of
the great fall. The best way to approach it, and to observe
the changes in the strata, is to pass up by the south side
of the stream. Here the usual chocolate-coloured Old Red
is seen, and in it a great bed, apparently not a dike, of
decomposing greenstone, with veins of calcspar and arragonite.
Gradually the sandstone becomes harder and more slaty, and
at length almost a quartzite. This is followed by a blue,
very hard flinty slate or quartzite, streaked with white, and
in some parts containing felspar crystals, succeeded in its
turn by a nearly white quartz slate, similar to that already
described (Art. 44). It has in many parts a blue tinge, as if
from transfusion of the colouring matter of the slate. This
quartzose slate is several yards thick, and is in contact with
the granite, here a hard and tough close-grained compound
of quartz and felspar — that is, a eurite — the mica, if occa-
sionally present, being in very small quantity. On the
south bank, over the junction, there is a mass of the ordinary
dark slate ; it shews twisting of the layers and banding with
blue and white, but is less altered than that in contact with
the gray quartzite in the stream. The metamorphism at this
junction is very remarkable; but the student must be
prepared to encounter many difficulties in his ascent of the
channel, and to submit now and again, with the best grace
he may, to a shower-bath from the dashing spray, as the
water bounds from ledge to ledge of the long fall.

One can easily pass from this point to the top of Goatfell
in three quarters of an hour; but we must now hasten to
our welcome rest at the pleasant Inn of Corrie.

118

EXCURSION VI.

TO THE NORTH SHORE.

59. THE coast section from Come to Loch Ranza is full of
interest. A long summer's day will be required to examine
it carefully — several days, if we stop to collect the fossils
which abound in some of the beds. The scenery, too,
is full of beauty — in parts bold and picturesque in the
extreme: we shall be tempted to stop very often to add
some new treasure to our portfolio. Leaving Invercloy
by the first steamer for Glasgow, we shall be at Corrie
for an early breakfast, and be ready to start fresh while
the day is yet young. But Corrie is for this, as for
several other excursions, an admirable point of departure-
A handsome and commodious hotel has recently been
erected here, and under its present management is most
comfortable. Loch Ranza is eight miles distant from Corrie
by the high road ; by the path we are to follow along the
shore, not more than between ten and eleven.*

The strata at the base of the carboniferous system, and
their contact with the old red sandstone, are well exposed
upon the shore. We shall trace them in ascending order,
beginning a quarter of a mile north of Corrie, at the
march of Achab farm, where the road bends towards the
N.W. This is the base of the series; but it is not a
well-defined base — there is, in fact, a gradual passage from
the Old Red system into the carboniferous strata. The
Old Red is here a conglomerate, and is overlaid by a

* Mr. Douglas, Postmaster at Corrie, son of the late Mr. Douglas,
whose great local knowledge was so freely placed at the servjpe of the
explorers of Arran, is well acquainted with the strata in the vicinity,
and with the geology and botany of the island.

EXCURSION VI. 119

limestone with imbedded pebbles, the same as those in
the conglomerate, forming a calcareous conglomerate.
This bed is followed by gray sandstone and concretionary
limestone, or cornstoue, consisting of red nodules, im-
bedded in shale. Other alternations of limestone and
sandstone follow, there being in all five or six beds
of limestone, the two lowest of which are seen rising up
the hill on the dip, S.E., over the conglomerate. The
whole series is to be regarded as perhaps rather the bottom
beds of the carboniferous series than the top beds of the
old red sandstone. There succeed these various beds of
sandstone and shale, till we reach an enormous vein or
dike of trap, which occupies the shore for more than 300
yards. In this are found basalt, greenstone, amygdaloid,
and concretionary trap, exfoliating in concentric coats; and
it is traversed by numerous veins of calc spar, steatite,
and quartz. The next beds seen are a red and a gray
limestone ; there is then a whin dike, south of which,
from the gate upon the road to the limestone quarry, the
shore is occupied by sandstone. The strata of limestone
are seen in the little harbour, and extend up the hill, rising
towards the north and dipping S.K at 36°. The rock
is of a bluish-gray colour, about twenty feet thick, and
consists of twenty-two beds of limestone, interstratified
with the same number of beds of red shale, the thickest
stratum seldom reaching one foot. The workings are
inclined adits descending in the line of dip. Over the
limestone are shales with hematitic iron ore. Several dikes
traverse the limestone, and alter its structure; fossils abound,
but are procured with difficulty. It has been remarked,
as indicating the tranquil nature of the deposit, that the
large producti uniformly rest with the convex side of the
valve downwards. In the sandstone upon the shore in
front of the village several common species of coal plants
are found. In the little bay under the hotel, and a little
south of Come, near the fall on Lochrim burn and the

120 GEOLOGY OF ARRAN.

freestone quarry, another bed of limestone occurs ; that
under Maoldon has been noticed already (Art. 58). The
many interesting trap dikes have also been noticed; and
those curious irregular ridges on the surface of the sandstone
mentioned as occurring on the Corregills shore, are even
more remarkable here. They sometimes stand up six or
eight feet above the surface, and have very sharp jagged
edges. In most cases they are independent of whin dikes,
and consist of matter originally less liable to disintegrate
than the sandstone, and probably introduced into fissures in
this rock after it had consolidated.

60. We return now to the contact of the carboniferous
rocks with the old red sandstone at the march of Achab
farm, and follow the latter rock northward along the coast.
Expanding inland, it rises into high cliffs. The forms
into which these have been moulded by the action of the
sea, the disposition of the natural wood upon them, and
the huge granite blocks which stand prominent by the
road-side, give a unique and most picturesque character
to this part of the coast. One of the largest blocks (on
the western edge of the road, among trees) we estimated
at above 200 tons; farther north on the eastern edge of
the road is another very large block standing upright on
its apex, perhaps let down so from a floating berg, or
originally imbedded in sandstone, afterwards worn away
as the tide ebbed and flowed around. The force exerted
by a strong man causes it to oscillate slightly as a rocking
stone. A romantic interest attaches to another boulder
south of the march of Achab farm. A garrison of eighty
men had been left in Brodick Castle by Cromwell ; against
these the natives became so irritated, on account of the
excesses of some of the soldiers, that when they were out
upon a foraging party in this direction, the Arran men
rose against them, and put them all to the sword, except one
young fellow who escaped, and hid under this stone. But
true love is ever fertile of expedients; and the Corrie

EXCURSION VI. 121

boulders proved no more a mystery to its promptings than
do the bolts and bars of a prison house — a faithful Arran
maiden, whose affections he had won, soon discovered his
place of concealment, ministered to him food and comfort, and
in a few days, forsaking her own people, escaped with him
in a fishing boat which had come from the mainland, and
was united to him on happier shores.

We cross the Sannox burn by a rustic bridge, where
it comes flashing down along its bed of granite sand, among
wild copsewood, through green shadows and gleams of sun-
light. Our path is now by the farm-house on the shore;
but we must not pass so near the finest view of Glen
Sannox without stopping for a little to look upon it. This
point will be readily found upon the road towards North
Sannox house. "VVe look down \ipon a broken foreground,
sloping on both sides towards the stream. On the right
the glen is bounded by the long steep ridge of Suidhi-Fergus
and the Castles; on the left by that of Cioch-na-h'oighe
and the prolonged ridge of Goatfell. Cior-Mhor stands
proudly up, closing the long vista. We return to the shore.
The high ridge of ground dividing North and South Glen
Sannox terminates on the shore, in a precipice called the
Blue Rock, from a decomposition taking place upon it,
due to the presence of iron and manganese, acted on by
trickling streamlets. Here, and along the shore, the strata of
Old Red still retain their southern dip, at a small angle. But
when we cross the North Sannox Water, we find the inclination
southwards much less; and as we advance a little, the strata
become horizontal. Still advancing, we find them dipping
in the opposite direction, at angles gradually increasing,
till the original dip of 15° or 20° is reached, but now
directed N.N.W., instead of S.S.E. The line from which
these opposite dips are thus directed is called the Anticlinal
Axis ; it is nearly in the direction of North Sannox Glen, and
of a remarkable advance of the granite in an easterly
direction almost to the bridge in Glen Chalmadale, not a

122 GEOLOGY OF ARE AX.

mile from this point The dips are shewn on the figure
annexed (fig. 21). The name was applied by Murchison
and Sedgwick; but the relations of the strata were first

Fig. 21.
Anticlinal Axis.

pointed out by MacCnlloch. It is obvious, then, that
the strata are successively newer as we advance along the
shore towards Loch Ranza. We shall pass over their basset
edges rising southwards, the dips being northerly, and the
inclinations on the shore from 10° to 15° and 20°; while
on the face of the hills above they are from 50° to 70°»
but in both cases alike towards the same point of the
compaM.

The Old Red here is cut by several dikes, of which
two are extremely interesting. One of these is about half-
a-mile north of North Saunox burn; it entangles a mass
of altered sandstone, wedge-shaped at either end, five yards
long and seven inches thick; the range is nearly north-
west. The other dike is at the angle of the shore, where
the Fallen Rocks first come into view, at little more
than half-a-mile distant The dike is best seen in the
sand under the grassy bank. The structure of the trap
varies, much of it being a fine blue greenstone or basalt
The change on the sandstone is remarkable. The interlacing
of the two rocks, and intrusion of string-like veins of trap
among the sandstone strata, as forcibly attest as any dike in

EXCURSION VI.

123

Arran the irruption of a liqiiid stream of lava into the
crevices of a fissured mass.

The Fallen Rocks are about two miles from Sannox;
they are an immense debacle of Old Red blocks, hurled from
a high far-up cliff; and now strewing the beach and steep

Fig. 22.
The Fallen Rocks.

slope in magnificent confusion (fig. 22). They seem freshly
fallen; yet Headrick described them sixty years ago as we see
them now.

61. At Groggan Point, immediately west of the Fallen
Rocks, we come on the lower beds of the carboniferous
series following conformably on the old red sandstone. They
are very similar to the corresponding beds at the march
of Achab farm, near Corrie; a calcareous conglomerate,
with beds of white limestone, being highest of the Old Red
series, or perhaps the lowest of the carboniferous.

These are followed, at the eastern side of Lagantuin Bay,
by sandstones dipping N.N.W., beds of red shale of moderate

124 GEOLOGY OF ARRAN.

thickness, and these by thick-bedded gray sandstone and
nodular limestone, till we reach the representatives of the
Corrie limestones, near the Salt Pans, prolific of the same
fossils. Lagantuin Bay, towards the lower portion of the
series, exhibits a great body of intrusive trap, in a position
almost at the base of the series, and thus exactly correspond-
ing to that already noticed a short distance N. of Corrie;
the two outflows being very nearly at equal distances from
the base of the carboniferous series. These eruptive rocks
consist of diorite of porphyritic structure (trap porphyry),
containing crystals of diallage, amygdaloid with zeolites,
steatite and carbonate of lime, trap tufa and claystones;
they upturn the shales to 58° from their usual inclina-
tion of 20°. Beyond and overlying these, as we advance
along the shore, there are various thick-bedded sandstones
with black and red shales, and beds of trap tufa; these
subside under the debris, and there is a considerable space
where no rock is seen till we reach Millstone Point, where
we again come upon thick white sandstones and gray
shales, dipping N. 5° E. at 39°. A little W. of this,
near the gamekeeper's cottage, we come upon one of the
most interesting spots upon this whole coast. Here, during
the autumn of 1865, Mr. Edward A. Wiinsch, Glasgow,
discovered that certain dark-coloured beds in the sandstone,
which had been only casually noticed by all previous
observers, and set down as sheets and dikes of trap, are
really hardened shales and beds of trappean ash, and con-
tained stems and branches of trees. A history of the
discovery and account of the beds, with a diagram, from
which our cut is taken, was published in the Transactions
oftfie Geological Society of Glasgow, vol. ii. p. 97 (November
1865). The beds were also carefully examined by Mr.
John Young of the Hunterian Museum, and Mr. James
Russell, coal surveyor, Chapelhall, whose zeal and great saga-
city as observers are known to all geologists; they were re-
examined by Mr. Wiinsch and ourselves in the following

EXCURSION VI. 125

summer: — The strata are exposed from a few feet outside
high water mark to the limit of low water, whence they
dip under the sea; the horizontal extent is about 150 yards;
the dip is K 10° E. at 39° to 60°. There are eleven distinct
beds of shales, trappean ash, and thin coal seams. The
ash beds are two to three feet thick, and enclose the trees.
The trappean ash closely resembles certain indistinct earthy
greenstones, with a semi-crystalline structure, but is really
an aqueous deposit, distinctly stratified and made up of
materials derived from plutonic sources, re-arranged by
water, and afterwards indurated by heat and pressure; and
is wholly different from the ordinary trap rocks of Arrau,
which are of much later origin than any of these beds, as
they cut through strata of much later age. But there
has been igneous action at all periods; and this case of
itself is enough to assure us that it was vigorous in the
lower carboniferous period. There must have been in this
neighbourhood a volcanic vent which showered forth ashes,
by which the trees in their growing position were successively
enveloped; while the existence of these trees on different
horizons shews long-continued periods of repose and vegetable
growth, and successive sinkings after each; so that the
vegetation out of which the coal was formed had time to
accumulate. Such frequent repetition of the surface of
growth, with intervening subsidence, is shewn in the coal
formation everywhere; and there are cases, 4n other parts of
Scotland, of beds very similar to these, in the lower car-
boniferous series, though without the remains of the ancient
forests so skilfully exhumed by Mr. Wiinsch. The trees
plainly grew upon a horizontal surface, which got after-
wards tilted up into its present inclined position, either
by the intrusion of the traps in the vicinity, or by the
upheaval of the central granite, which, we have seen,
may have occurred after the lower carboniferous beds were
deposited.

The beds are shewn on the annexed diagramatic section
(fig. 23) in the position, and of the terraced form, into which

126

GEOLOGY OF ARRAN.

they have been worn by the action of the waves. The
beds a a are the hard trappean ash, two to three feet thick,

6 a

Fig. 23.
Trappean Ash Beds.

enclosing the trees; b b, seams of coal, shale, and ash, in
alternate layers ; c c, the stems of trees, sending down roots
into the bed below. " Twelve or fourteen trunks in all
have been observed; roots descend into the sliale, and
branches lie about in the beds, fully mineralized, yet having
the structure beautifully preserved. The large stems are
Sigillaria, with five or BIX. internal piths shewing the endo-
genous structure ; the smaller branches have been identified
as belonging to Sigillaria, Halonia, and Lepidodendron; also
AntholitJies and seed cones, Splienoptiria, and Stigmaria
fieoides (root of Sigillaria) have been identified. The
structure of the plants indicates that they grew in a marshy
situation; and the fine ash from the volcanic vent quietly
and gradually enveloped the stems." *

The seed cones here referred to have been very recently
described and figured by Mr. Binney of Manchester (Tr.
Palaemit. Soc., vol. for 1870). They are known to be the

* Tr. Qeol. Soc, Ql, vol. ii. p. 97-8.

EXCURSION VI.

127

fruit of Lepidodendron, and are named Lepidostrobus. There
are two other species in the ash beds, L. lotus and L.
Wuenschianus. One of Mr. Binney's figures, L. ambiguus,
exhibiting the internal structure of the cone, with portion of
the stalk attached, is reproduced with permission in our
woodcut (fig. 24).

Fig. 24.

In the spring of the present year there was found in these
beds a palatal tooth of the Ctenodus cristatus, a species of that
typical genus of fossil fishes which gives its
name to one of the four great divisions of
Agassiz, from the structure of the scales re-
sembling the teeth of a comb in their marginal
divisions. The structure of the scale and
tooth is represented on the annexed sketches (figs. 25, 26).

Westwards of these beds, and of course overlying them

Fig. 26.

there succeed sandstones and shales, thin courses of fel-
stone or porcellanite occurring in the red shales, and
beds of limestone with Producti, species of Echinodermata

128 GEOLOGY OF ARRAN.

and other fossils, alternating with these and with the sand-
stones. The whole series dip more to the west and at
higher angles, often as great as 70°. The total thickness
of the beds W. of Millstone Point, as far as the Salt Pans,
may be estimated at 5000 feet, of which the limestone and
shale make about 150 feet.

Sandstone and shale with seams of coal succeed these
till the beach is obscured by shingle. The principal coal
seam is three or four feet thick; but has been worked out,
as far at least as the level of the sea, which now invades
the workings. It was used only at the adjoining salt and lime
works, and never exported. There were workings on two
or three seams in the direction of the dip, which is nearly
N. at 45°; but all were abandoned when, from the
depth of the workings, the sea gained access to the pits.
The coal is of that variety called blind coed, containing
an unusual quantity of carbon, and burning without flame
or smoke. Many vegetable impressions, chiefly ferns and
calamites, were found in the coal below the shale; the stems
of the latter being formed of a hard coal with a bright
shining fracture. This character of the coal may be partly
due to the action of several dikes which here cut the strata.
An adit was opened in the hill above on the strike of the
seam, but no coal was found. Still, there seems little doubt
that there must be considerable beds of coal yet untouched.
The shale and coal tract is bounded by a black limestone
below and red limestone above, 1400 yards apart, and
extending up against the schist precipices, thus affording
ample room for a considerable development of the coal
beds.

A little to the west of this, beyond a shingly beach, sand-
stones and shales again appear. The shales contain beds of
ironstone, some of which have the structure of septaria; and
these, with the variegated shales, form on the shore a flat
platform, with a tesselated appearance, like a mosaic pave-
ment. " I doubt," says Mr. Headrick (p. 210), " if the most

EXCURSION VI. 129

skilful mason, or even a mathematician, could produce any-
thing more regular or more beautiful."

Farther on — about half-way between the Salt Pans and
the " Cock of Arran" — several beds of red limestone, rich in
fossils, and of red shale, occur. It is interesting to notice
the perforations of pholades in the limestone above the
level of the present tides, as being a striking collateral proof
of that change of level to which we have so often alluded.
The beds are so numerous, and the alternations so much
like those already noticed, that a description of them would
be tedious; but the variety of the fossil contents will amply
repay a careful examination.

These beds are succeeded by a series of sandstones, shales,
and fine conglomerates, overlaid in their turn by variegated
marls containing nodular ironstone, and by white sandstone.
Over the latter, a little way east of the Cock, lie beds
of fine red sand, alternating with fine conglomerate, the
dip of both being K 65° W., at 23°. These beds, with
the white and red sandstones which succeed them westwards
till the schist is reached beyond the Scriden, are the
uppermost members of the whole series, which begins to
overlie the old red sandstone near the Fallen Rocks. From
their position and mineral character they have been classed
as new red sandstone by Murchison, Sedgwick, and Ramsay.
Mineralogically, they have a much greater resemblance to
lower permian strata than any of the rocks in the southern
district; but we refer them, notwithstanding, to the upper
carboniferous series, to which they are more closely related
by their fossils.

The Cock of Arran, near which this series begins, is a
large isolated mass of sandstone, resting on the beach, a
noted landmark among sailors. When seen in front from
the sea the block had the form of a cock, with expanded
wings, in the act of crowing. The resemblance is now less
striking, as the head has been broken off'. Beside this block
there are two singular whin dikes close together, on the

130 GEOLOGY OF ARRAN.

flat beach, about two feet wide. They terminate in the
sandstone, almost opposite to one another, and are prolonged
in contrary directions.

62. The Scriden is a headland whose base is strewed with
immense masses of sandstone. These fell about 100 years
ago with a loud noise, heard in Bute and Argyll. The
debacle is more extensive than that of the Fallen Rocks,
but inferior in grandeur. It was produced by a landslip of
the mountain side, the traces of which yet remain in a long
deep rent near the summit. Scrambling for a long way
among the fallen masses, we reach an open shingly beach,
along which the line of the old slate, advancing from the
interior, strikes the shore, and cuts off the red sandstone;
but the nature of the ground does not permit the junction to
be seen. The slate here dips about S.E. at 40°. The change
takes place near a glen, with a burn called Alt-Mhor (large
burn). The variety called chlorite shite occurs here, and
quartz abounds in veins and beds in the slate.

A short distance forward, at Newton Point, where the
coast bends round into Loch Ranza, and a small stream,
called Alt-Beith (birch burn), enters the sea, there occurs
one of the most instructive sections to be seen in Arran.
Strata of sandstone again occupy the shore for 300 yards,
dipping into the sea, and resting along the platform in front
of the cliffs upon the upturned edges of the strata of slate.
These make an angle of 40° with the horizon, and dip 40° E.
of S., that is towards the granite of the nucleus ; the strata
of sandstone are inclined at 25° and dip 55° W. of N. ; the
dips being thus nearly in opposite directions (see Art. 5,
sub fin). This uneouformability indicates that the slate,
itself a sedimentary deposit, had not only been formed in this
regular stratitication, but had undergone a general disturb-
ance before the sandstone beds were thrown down upon it.
The position of the slate-strata is thus not directly related to
the granite; the dip and strike are really related to the
great axis of elevation traversing the mainland towards the

EXCURSION VI.

131

German Ocean. This position had been acquired before the
Arran granite was injected amid the strata of slate, and ere
yet any of the sandstone beds which succeed it had been
deposited. These sandstones (old red and carboniferous) are
conformable to one another, and the deposits blend at both
sides of the section, Achab farm and the Fallen Rocks. But
we see here, as repeatedly noticed already, that neither
deposit has any relation to the stratification of the slate,
which had sustained extensive dislocation before the deposit
of the old red sandstone had begun. The cut annexed
represents the appearances at this place.

The sandstone strata here alternate with beds of limestone

Fig. 27.

a, slate ; b b, "beds of white siliceous limestone ; c, sandstone
and conglomerate.

and conglomerate, as shown on the figure annexed (fig. 27).
The lowest bed, b, next the slate, a, is a hard crystal-
line white limestone, about six feet thick ; it contains
quartz pebbles, schist, and diffused siliceous matter, and
is without fossils. There are several beds higher in the
series, but the thickness is less. The total thickness of the
various beds is not more than fifty feet. Whin dikes tra-
verse both the slate and sandstone. The presence of this
peculiar limestone fixes the age of the deposit, which
resembles in all respects that already noticed at Achab farm
and the Fallen Rocks, as forming a gradation from the Old
Red to the carboniferous series — there being a difficulty in
determining to which it ought to be assigned. We are
inclined to the view that it is the lowest member of the
carboniferous system. The occurrence of the limestone here

L6'2 GEOLOGY OF ARRAX.

is noticed by Jamieson (Min. of Scot. Isl., vol. i. p. 78, edit.
1800), from whom our cut (fig. 27) is taken; by Headrick
(p. 206, 1807); and by MacCulloch (vol. ii. pp. 376, 356,
1819). Later writers have classed the beds as new red sand-
stone, overlooking the limestone, and have given the dips
erroneously. The deposit is a mere isolated patch, and has no
connection with the sandstones eastwards, which abut upon
the slate, and are clearly the uppermost members of the
carboniferous series in this part of the island.

Very few granite blocks occur along the north shore, and
none of more than two to three tons in weight. But as we
enter Loch Ranza and its glen, opening directly on the
granite nucleus, we meet with boulders of great magnitude.
Two of this kind, with many lesser, rest on the beach at the
north-east angle of the bay. These are of the coarse variety,
and their position here seems to prove a transport from the
nucleus; that the ice stream had not its whole issue south-
wards, but discharged its load of blocks in this direction also.

63. One generally reaches this point late in the day, when
the rays of the declining sun already fall strongly on the
outer part of the bay, the upper portions being in the deep
shadow of the high western hills. The old castle on its
raised bank of shingle, which has kept its place steadfastly
against wind and tide for at least 2000 years, is a fine object
in the foreground. The middle of the picture is occupied
by boats, in from the fishing, and by nets suspended between
high poles, beyond which is a line of white cottages on the
west side, and houses of higher pretensions on the east, from
which the curling smoke rises up invitingly. The back-
ground is a dark circle of gloomy hills, now reflected in the
lipping tide, which give an air of peculiar solemnity and
seclusion

" To the lone hamlet, which her inland bay
And circling mountains sever from the world."

Beyond, and over all, rises the serrated granite ridge of

EXCURSION VI. 133

Castle Abhael and the Sui, gleaming now in the golden light
of sunset, and contrasting strongly with the dark slate
hills which encircle the bay. The place altogether has
a singulai-ly picturesque and unique aspect; the stranger
will say he has never seen such a hamlet and bay before. Yet
will he find most comfortable quarters at the unpretending
inn, and even luxuries not easily found elsewhere.

The stranger who has got but one night at his disposal to
spend in Loch Ranza, should contrive to reach the place so
early that he will have time for a stroll along the sea-shore,
to the west, in the direction of Glen Catacol. Towards sun-
set, on a clear and still summer evening, the scenery appears
to the highest advantage. It has then a quiet but touching
beauty, which steals into the very soul.

There are several objects of much interest about Loch
Ranza. The castle is a regular structure of stone and lime,
with thick walls, and is still in tolerable preservation, though
now roofless. There are several small apartments and a
large hall reached by a narrow stair. It must have been a
place of great strength before the use of artillery was known.
It stands on a bank of shingle, running across the mouth of
the bay, and forming the harbour. "We first find it men-
tioned as being a hunting-seat of the Scottish kings in 1380;
it was then reckoned one of the royal castles. But the date
at which it was erected is not known. In 1452 the lands
and castle of Loch Ranza, with the lands of Sannox, were
granted by the Crown in feu to Lord Montgomerie ; Ronald
M'Alister, who held them before, having refused to pay rent
on a plea not deemed sufficient. The Montgomeries of
Skelmorlie, ancestors of the Eglinton family, got a grant of
the property in 1685, with much of the north end of the
parish. They, however, lost the entire property, as an
unredeemed mortgage for £3600, to the Hamilton family in
1705. The Duchess Anne, then the head of that family,
built a chapel and established a missionary station at Loch
Ranza, for the maintenance of which she mortified a sum of

134 GEOLOGY OP ARRAN.

money, producing £27 a-year. No trace of the chapel now
remains. The ruins of the convent of St. Bride, which till
lately existed here, are also swept completely away. Some
remains indicating the situation of an old cemetery have
lately turned up near the supposed site. The only interest
connected with it lies in association with the well-known
scenes in Scott's Lord of the Isles, in which the scenery of
" fair Loch Ranza " is very prettily and graphically de-
scribed. The harbour is the resort of fleets of fishing boats.
It is a most picturesque and exciting scene to witness the
launching of these boats on a summer evening, under a
favouring wind. In rapid succession they drop down from
the harbour into the outer bay, and dart away in bounding
glee, some out north into Loch Fyue, and others away to the
left, across the broad waters of Kilbrannau Sound. Though
so land-girt and safe-looking, Loch Ranza is by no means a
safe anchorage. The most fearful squalls, sometimes even in
summer, fall upon it down the narrow opening of Glen
Eas-an-Bhiorach, between Torneadaneoin and Meal-Mhor.
The currents of a S.E., S., or S.W. wind traverse the whole
length of this long glen, hemmed in between high hills, and
issue from the narrow gorge with terrible violence. Vessels
are thus often driven from their moorings, and obliged to
seek safety in the open frith, or by running for Loch Fyne.

135

EXCURSION VII.

BY THE WESTERN SHORE.

64. THREE interesting excursions may be made from Loch
Ranza — one across the granite nucleus, one by the western
shore, and a third back to Corrie. From noiie of them,
however, will it be easy to return to this place on the same
day; they may therefore be taken at different times, after
an interval. Most tourists will prefer the last — a return to
Corrie by the inland rout.

A walk by the shore, from Loch Rauza to Dougrie, about
eleven miles, will more amply reward the lover of the lonely,
the wild and picturesque, and the botanist, than it will the
geologist. Still there is the slate to study ; the whole walk
is upon slate; and the many changes in its mineral character,
the effects of former sea action upon it, as shewn in the
ancient cliff, which in some places advances boldly on the
shore, and in others retires inland, and as compared with the
like action on the sandstone, will form interesting subjects of
inquiry. Abrupt changes in dip will be noticed in the
slate, at the north side of Glen Catacol. With the exception
of a few singular inversions, due probably to dikes, it
maintains a pretty uniform dip and inclination, S. 40° E.
at about 40°, till we approach Glen Catacol, north of which
an anticlinal seems to occur, owing to an advance of the
granite in the direction of the bay. On one side the slate
rises up towards the granite, as if thrown off from it, the dip
being 1ST. 65° W. at 53° ; on the other it dips E. and S.E.,
at about 30° to 40°. But this dip is not maintained : it is
sometimes directed towards the granite, in other places
" end on " to it, or southwards, and in others again off the
nucleus, or nearly west ; but there is not observed here that

136 GEOLOGY OF ARRAN.

mantling stratification around the nucleus which is usually
seen in granite tracts. — The entrance of Glen Catacol
exhibits some fine terraces. On the south side there are two ;
one about forty feet in height, with which a mound rather
than a terrace on the other side corresponds; and a lower ter-
race about twenty to thirty feet high, which reaches down
to near the sea. "With this a terrace on the north side
corresponds, beginning near the farin-house, and sweeping
round eastward. The front of the higher terrace is cut into
promontories and bays, indicating the action of water upon
it after its deposition ; all the terraces consist of alluvial
matter, not of rock. When the sea stood at its former level,
so often alluded to already, the plain in the centre of the
glen must have been entirely under water — there is but a
thin soil over rolled gravel through all its central part. On
the north-east part of the glen, and out eastwards over the
mountains towards Torneadaneoin, some very interesting
junctions are to be seen ; but we must pursue our way now
towards Dougrie.

It is only on this extreme north-western border of the
slate tract that the micaceous character which prevails in
Cantire is assumed by the slate. Near the Free Church,
which stands on a lovely picturesque spot in North Thunder-
gay, a fine mica slate, banded with quartz, occurs ; east of
this there is much chlorite slate, and one mile west of it,
near the secluded cemetery, blue argillaceous slate, mixed
with the chloritic variety. But generally there is on this
side a greater tendency to the micaceous character. On the
Imochar shore, and towards Dougrie, the ordinary character
has become again established. The slate ceases, and the old
red sandstone begins to form the front of the cliff a little to
the west of the mouth of the lorsa ; but the junction is not
seen. The stratification appears to be conformable. There
are comparatively few dikes on this coast A little east of
the Free Church there are two, each of which appears to
bifurcate. On the Pcnrioch shore a great bed of greenstone,

EXCURSION VII.

137

100 feet broad, occurs. We did not notice a single bed
either of porphyry or claystone. Many bouldei-s of granite
occur ; northwards they are of the coarse-grained variety,
but of both varieties on the South Thundergay and Penrioch
shores. A few were estimated at as much as twenty to
thirty tons weight.

As the coast from South Thundergay southward is some-
what monotonous, we would recommend the tourist, if the
day be clear and steady, and he has had an early start, to
diverge inland at South Thundergay, visit the lovely and
secluded corrie and tarn, called Corrie-an-Lachan (hollow of
the lake), at the north-east of Meall-Biorach, ascend Ben-
Bharraiu from the north, pass along its eastern ridge, and
descend upon lorsa water-foot, through Glen ScaftigilL He
will be delighted with the wild scenery, and will notice many
objects of geological interest.

The terraces already noticed (Art. 18) at the mouth of the
lorsa, in front of the Duke of Hamilton's shooting lodge, are

Fig. 28.

represented on the annexed woodcut (fig. 28) ; they are the
most remarkable in Arran. They are the remains of a
vast accumulation of detrital matter which once filled
the whole of the deeply-embayed area at the mouth of
the river, and was most probably deposited as a terminal
moraine to two glaciers, which moulded the sides of
Sail-Chalmadale, and united their streams near the head of
the bay. The seaward front and inland extension would be

138 GEOLOGY OF ARRAN.

originally nearly half a mile each. The detritus backs
against the sides of the sloping hill all around, and is seen to
rest against the natural rock, fronting the sea, outside the
bay, at the same level with it, and worn like it into the
existing cliff. This effect is especially remarkable on the
west side. Here the slate, the old red sandstone, and the
detritus form, in succession towards the east, the front of a
cliff, at first of nearly uniform height, and then gradually
declining eastwards, all the three portions presenting exactly
the same evidence of sea action. They form, in fact, an un-
broken line of old sea cliff, carved out on all three alike
when the land was rising, in the manner already pointed out
(Arts. 18-20). If the terrace existed before, and the
moraine was thrown down upon it, then there had been an
earlier wearing of the rocky front, the evidence of which
should still exist at the back of the detrital mass, and also
under it on the floor of the terrace. Boulder clays, we know,
are in some of the glens accumulated against a rocky face
worn earlier, and scratched rocks are seen xipon the floor of
the terrace.

The contact of the slate and sandstone is not seen in the
cliff's, nor on the terrace in front, where worn masses of
slate in front of the slate cliff, and of sandstone in front of
the other two portions, all in situ, rise up through the
shingle. The foundation rock all round eastwards by the
lodge and east side of the bay is sandstone. The detritus
was cut into the existing forms, by the joint action of the
sea and river ; the mounds present in front rounded promon-
tories and semi-circular bays, and on the ascent terraces at
several successive heights. An isolated mound in front of
the lodge was removed in order to open up the view east-
wards. Among the detritus there are beds of fine sand,
much sought after by gardeners, and often carried to the
mainland.

There is no hotel in this part of the island, but very com-
fortable quarters can be had at a farm house in the hamlet of

EXCURSION VII. 139

Auchincar, a short distance east of the lodge. The tourist
can then easily find his way to Brodick, or he may pass along
the shore by Mauchrie water-foot to Tormore and King's
Cove. From Dougrie to the Tormore pitchstones, or to the
Tormore Stone Circles, is a good hour's walk. Ben-Bharrain
appears to best advantage on this walk ; and we often pause
upon our way to turn round and gaze on its stupendous mass,
and to wonder how it came there. A quarter of a mile from
the shore there is a wooden footpath across the Mauchrie.
About half a mile from the sea, between the farm houses of
Mauchrie and Tormore, carboniferous limestone, with pro-
ducti and encrinal stems, occurs as a bed in the sandstone of
Mauchrie water, and is very similar to the Come beds; and,
like them, the limestone is not far from the base of the series.
In Mauchrie burn, close to Mauchrie farm house, the Old
Red beds are seen ; but no precise line can be laid down for
the boundary. Glen lorsa opens on this part of the coast ;
and here the number of granite boulders greatly increases —
the proportion being now in favour of the fine-grained
variety, which spreads widely over this glen, and reaches to
a great height on both sides of it. From lorsa water-foot the
coast section presents first old red sandstone and conglom-
erate, and then carboniferous sandstones. We observed
scratched boulders about Auchiucar, and masses of old red
sandstone, which in this situation are not travelled, also
striated.

140

EXCURSION VIII.

BY THE INTERIOR OF THE NUCLEUS.

65. A walk across the interior of the granite nucleus, from
Loch Ranza to lorsa water-foot, will be good work for a long
summer's day; or the geologist may return from Auchiucar
to Loch Ranza by this route. We preferred the former
course, passing up Glen Eais-na-vearraid by the base of
Torneadaneoin. The granite junction here is very cele-
brated, having been often referred to in the discussions be-
tween the abettors of the igneous and aqueous theories — the
Huttonians and Wernerians. In our view, the appearances
are strongly confirmatory of the intrusive character of the
granite. About a mile south of Loch Ranza we leave the
road and pass up to the base of Torneadaneoin, keeping on
the west side of the stream. Torneadaneoin is a bold and
finely-shaped mountain, 1057 feet in height, which rises on
the east side of the gorge through which the fierce blasts
already alluded to often come down, while Meall-Mhor rises
over it on the west.* Through the gorge a torrent descends
into the head of the loch, having cut a deep channel through
the hard layers of slate. The place is wild and picturesque ;
at some points great caution must be used in selecting our
footsteps. It was through this gorge that a glacier, which
filled the wide mountain glen beyond, debouched upon the
shore of a former period, and threw down into the bay the
mass of detritus, which shoaled the water in this sheltered
situation, and made the after-work of the river comparatively

* Torneadaneoin — the mountain of birds' nests ; Meal or Muil-
Vawr— the great, round-topped mountain ; Moil, Muil, whence Mull
— is a round head, and is applied to mountains and headlands. The
Gaelic language does not recognize diphthongs.

EXCURSION VIII. 141

easy. The river and tides have re-arranged the materials,
and rolled them into a perfect marine shingle; but doubtless
this was the origin of much of the debris which has filled up
so large a part of the bay. The shingle bank on which the
castle stands has very much the appearance of a terminal
moraine; and it is just in such a position that a glacier like
that supposed would for a long time throw a moraine down.
There are traces of a former terrace along the sides of the
bay, backing up against the mountains, in the position which
lateral moraines would occupy. When we said that the
shingle bank has kept its place for 2000 years, the reference
was to the establishment of the existing levels, not later than
the time of the Roman invasion.

66. At the upper end of the gorge, where the open glen
begins, we are upon the junction in the river bed. Many
fine branching veins of granite run into the slate, narrowing
to threads as the distance increases, the granular structure
becoming at the same time more minute; in some places
granite bands are interstratified with the slate. The slate is
also penetrated by quartz veins, of which some are parallel to
the layers of slate, and others intersect them at various angles ;
in the former case, they conform to all the contortions of the
slate. In some of the larger veins the granite is coarse-
grained ; but usually a more compact or confusedly crystalline
strip separates these from the slate. The appearances are
finely exhibited in the front of the mountain, which is cut
steeply down, and shews the veins in section. It is this
circumstance which renders the junction here by far the
finest in the island. " The whole mountain," as Murchison
and Sedgwick express it, "abuts against the granite, which
moulds itself into the broken edges of the slate, and runs
into it through the gaps and fissures." Large veins are seen
to emanate from the granite below, and pierce the slate in a
slanting direction towards the summit. There are also many
fine veins and numerous alternations where the two rocks
approach; and here the usual changes are well exhibited —

142 GEOLOGY OF ARRAN.

nowhere is metamorphic action more decided. The whole
neighbourhood, indeed, is well worthy of careful study. It
will be found very instructive to ascend the hill along the
plane of contact, and examine the various junctions. Many
appearances equally striking are also to be seen across the
hill-tops between this point and Glen Catacol, affording
ample scope for a separate walk. We may ascend from the
Torneadaneoin junction, and return from Catacol by the shore;
or first visit Catacol from Loch Ranza, examine the terraced
ridges at its mouth, and then ascend by the granite junction
at the north-east angle of the glen. The descent upon
Loch Rauza from the high platform is not difficult. The
walk affords many grand views, particularly late in the
afternoon.

67. From the Torneadaneoin junction southwards the
walk is wholly on granite, and its chief interest, apart, of
course, from its picturesque attractions, consists in examin-
ing the two varieties, the coarse and fine-grained, their
relations and respective limits, and the dikes which traverse
them. Few walks, indeed, in Arran, are more delightful on
a fine day : the peaks and ridges are seen in new and grand
aspects, enhanced by the solemn stillness and desolate char-
acter of the scene. The silence is broken only by the wild
scream of the curlew roused from her nest, or the answering
call of her mate aloft, by the whirr of the startled wild-fowl,
or sudden bound of the red deer from his lair. From Loch
Ranza to Dougrie there is not a single human habitation, not
even a herd's shieling; and during the ramble of a long sum-
mer's day we shall meet only the wild denizens of the rock
or moor. The red deer seek chiefly the shelter of the lower
valley of the lorsa ; and it is thence, by the wild heights and
rocky knolls towards the watershed, that the sport of
" stalking" is occasionally followed. The deer are not very
numerous; but they have been put into some other covers,
and the small American species has lately been placed in the
woods about Brodick Castle.

EXCURSION VIII. 143

The botany of this district is meagre, as is usually the case
with open uniform granite tracts; the only good plant is the
rarer species of Pyrus, which is found in rocky places on the
stream and at the base of the castles. The junction of the
fine granite with the coarse generally takes place where the
gradual slope ends, and the steeper climb to the mountain
tops begins. An important service would be rendered by
tracing this line along both sides and observing the appear-
ances— whether they are similar to those already noticed by
iis on the south side of the district (Art. 14); whether veins
everywhere pass from the fine into the coarse. It would be
almost necessary to " camp out" two nights running, and it
is not easy to accomplish this in a climate so uncertain. On
the hill-sides N. of the watershed, Mr. "Wiinsch has recently
noticed that there are two terraces corresponding on opposite
sides of the glen, strewed with blocks of fine granite resting
on coarse, which have a striking resemblance to lateral
moraines.

Having reached the summit-level on our ascent from the
north, we are at the curious little tarn Loch-au-Davie, which
discharges both ways, as already noticed (Art. 2.)

In descending the lorsa about a mile, we pass off the fine
granite and come upon the coarse variety, at first in the bed
of the stream and then in both banks. Near the waterfall
a basaltic dike, twelve feet wide, crosses the bed of the stream
nearly at right angles, and intersects both granites. It is
indicated on our section (fig. 3), and was traced by us a long
way in both directions ; of many others there are indications
along the hill-sides below Loch Tanna, and on the walk to the
north-east of Sail-Chalmadale ; but the disintegration of the
dike and granite alike is so rapid that the width cannot be
ascertained with any exactness. In tracing these continua-
tions we must mainly look to the direction. In later visits
to the lorsa valley I felt satisfied that this range is from side
to side of the valley, and that the dikes which ascend from
the depths of Glen Rosa and cross the Ceims, notching their

144 GEOLOGY OF ARRAN.

crests deeply, are the same dikes which were here seen to
intersect the bed of the lorsa and ascend the ridges to the
west — to emerge from the fine granite which covers the
lower slopes and again traverse the coarse variety on the
higher summits. There are many routes now open to us by
which to complete our day's walk. We may diverge to the
right and examine the great deposits of decomposed granite
at the head of Glen Catacol — which we once hoped would
be found suitable as a china-clay, but which turned out to
be too coarse and gritty — then ascend Ben-Bharrain, pass
along its horse-shoe crest, and then, having crossed the
junction of the granite with the slate, which here ranges
W. 25° S., or nearly "end on" to it, dipping S. 25° E. at a
high angle, descend upon Imochar; or we may pass on
southwards by the base of Sail-Chalmadale to the mouth of
the lorsa, the two routes here uniting at Dougrie. We are
then near comfortable quarters at Auchincar ; or, by previous
concert, a vehicle may be in waiting to carry us to Brodick
— ten miles — very welcome it will be after so long a march.

If the day is bright and yet young, we may lengthen our
walk pleasantly by passing from the waterfall to the summit
of Sail-Chalmadale, examining the junction of the coarse and
fine granites, and the veins of the fine in the coarse, which
are well seen on the southern front — thence we may descend
again into the valley of the lorsa, mount the eastern ridge
and examine whin dikes which traverse both granites, in
the gullies which furrow its sides, pass upwards to the
plateau on which Loch Ghnuis is situated, examine the
singular junction there, already described, and descend to
Brodick either by the Garbh-Alt and Glen Rosa, or cross
the wild moors south-eastwards to the summit level of the
" String." It will be a good day's work — one whose perform-
ance more than once we know to be possible without serious
fatigue.

145

EXCURSION IX.

LOCH RANZA TO CORRIE.

68. THE inland route from Loch Rauza to Corrie, a distance
of nearly eight miles, presents a few objects of interest. A
walk of three miles by the high road, which winds screw-like
along the hill-sides, by the edges of the deep glens, brings us
to the summit-level of Glen Chalmadale and the watershed
into North Glen Sannox. The views of the northern front
of the ridge of the Castles and the Sui are very grand from
this point. A few hundred yards below the summit, as we
pass down into Glen Dhu, granite appears by the road-side ;
and in the river bed, a little farther, a junction of the slate
and granite is very well exposed. A glance at the map will
explain the occurrence of the granite here. The principal
junction is at the point where the river comes close in under
the bank, along which the road passes. Here there are
some peculiarities well worthy of notice. The two rocks are
seen to come close up against one another, most distinctly
contrasted ; but there are no veins emanating from the
granite. This rock is coarse-grained, much coarser than in
any other junction we have noticed; yet it is more compact
than the coarse-grained granite usually is, and, though hard,
brittle. About one hundred yards down the stream a vein
of fine compact granite, five feet broad, traverses the slate;
and thirty yards farther, another vein, one foot broad, bluish
coloured, fine-grained, and very brittle. The entire mass of
slate between this lowest vein and the main junction above
is much altered : it is extremely hard and tough, contorted
in the laminae, and, being much veined with quartz, is
assimilated in aspect to a granite rock. Below the lowest
vein it speedily assumes its usual appearance. Mr. Ramsay

146 GEOLOGY OF ARRAN.

was the first to notice this interesting junction; and it is
well described by him. Several other junctions may be seen
in the branches of the main stream.

A little farther down the glen, and less than one mile
from the bridge, some well-marked glacial striae are seen on
the road-side, upon a mass of slate in situ. They cross the
laminae, and are directed nearly due east and west. The
slate also is rounded off. On the hill to the north, some
great blocks of coarse granite rest upon the slate. These
travelled blocks reach a good way up the south front of the
coast range; but in crossing, on one occasion, along the
summit, keeping generally near the edge of the cliffs, we did
not find a single block. It would seem, therefore, that the
few, and comparatively small blocks, on the north shore,
must have worked round, from either side, under the force
of the waves. Slate was found everywhere on the summit,
and far down upon the steep fronts, where the junctions take
place. Some curious varieties of slate will be observed in
the river bed, on both sides of the bridge. A short distance
below the bridge we leave the slate, and after an interval
come on coarse conglomerate, no junction being visible.

Many years ago, a large population, the largest then
collected in any one spot in Arran, inhabited this glen, and
gained a scanty subsistence by fishing and by cultivating
fertile plots on the sunny hill-sides. In 1832, the whole of
the families, amounting to 500 persons, were obliged to
leave the island, but were furnished with the means of
reaching New Brunswick. They formed a settlement at
Chaleur Bay, which became very prosperous. Garden en-
closures, tree-clumps, solitary fairy thorns, and ruined wall-
steads, still remaining, give a melancholy interest to this
secluded glen.

Near the summit level of the road between North and
South Sannox, and a little to the north of Sannox House,
the sandstone strata, on the east side of the road, are marked
with striae, rendered somewhat fainter than those on the

EXCURSION IX. 147

slate by the effect of decomposition. Like these they are
transverse to the laminae, and are directed W. 4° S. — that is,
in the direction of the ridge of •Suidhe- Fergus. If we assign
their origin to ice, then we must admit that the moving
masses advanced in parallel courses, in both cases, from high
ground on the west.

We are now within a few yards of the spot where the
finest view of Glen Sannox is obtained, and to which a
former excursion brought us. Taking another survey of the
grand scene in its evening aspects, we press forward to
Corrie, to catch the Clyde steamer on her way to Brodick.

148
EXCURSION X.

TO HOLY ISLE.

69. MAKE choice of a bright and quiet day, and leave
Invercloy for Laralash by the mail steamer from Ardrossan.
On reaching Lamlash, a boat must be hired to carry you to
Holy Isle. The pier having fallen to pieces, the steamer
does not stop, and there is no one permanently resident who
could be signalled to come off. A few hours will suffice to
examine the island, and the charge for a boat is moderate —
from 6d. to Is. per hour. The basis of the isle all round
is sandstone, which rises to the height of 100 to 150 feet;
the rest of the mountain, to the height of 1030 feet, is
composed of claystone, so that this rock has the thickness of
about 850 or 900 feet. On part of the east side, however, it
is much less than this, as the sandstone rises there much
above its usual height. The rock is of igneous origin, a
member of the trap family, and varies in structure from a
soft claystone to a compact felspar; the harder varieties are
called clinkstone. As in most traps, the prevailing form is
prismatic; and on the east and south-east sides of the island
columns of great length appear in the high precipices. But
the schistose structure is also common; and, as in cases
already mentioned at Corriegills, both structures occur in the
same mass, the slaty fracture being at right angles to the
axis of the prisms; the ends of the prisms first divide into
laminae, and the mass gradually assumes the slaty structure.
The weathering of the rock is remarkable, and extends to a
considerable depth, presenting successive concentric zones of
different colours, which have a very pretty appearance in many
specimens. The slaty structure itself, in the case of this
rock, seems to be but a step in the process of decomposition.

Various masses of greenstone occur, both as dikes and

EXCURSION X. 149

beds; one on the south-east side of the island presents
remarkable alternations with the sandstone, and alters its
structure. The felspar rock itself is seen in one place on the
east side to intersect the whole body of strata, and to be
connected above with the overlying mass of the same rock.
Here, also, there is a very interesting vertical vein or dike,
intersecting an oblique trap vein, and shewing in the centre
fine black basalt, containing zeolites and glassy felspar; on.
each side of this there is a lamina of a black substance, inter-
mediate between basalt and pitchstone, and at the outer
surfaces a coating of vitreous pitchstone one-tenth of an inch
in thickness. These substances, indeed, are all of one origin,
the variations found here, as well as in other places, being
due merely to different rates of cooling in the once fused
mass (see Art. 52). The same oblique vein is intersected by
a vertical vein of concretionary trap. Dr. MacCulloch gives
a drawing representing the appearances (West. Isl., vol. iii.
pi. xxiv. fig. 2).

The granite blocks on this island have been already noticed
(Art. 19). On the west side, near the landing, there is a
raised beach or shelly deposit thirty feet above the present
sea-level. Here also once stood a fortress, built in the
middle of the twelfth century by Somerled of the Isles,
of which no trace now remains. Adders abound upon the
island, but their number is said to be diminishing.

70. The tourist must visit the hermitage of St. Molios.
It is situated a mile south of the landing-place, and is a
natural excavation in the old sea-cliff. The conjectural
account of St. Molios usually given is quite inaccurate. The
following is the correct legend, kindly made known to me by
the late John M'Kinlay, Esq. of Bonnington, near Edin-
burgh, formerly of Rothesay *: — " The legend of St. Lasrian,

*Mr. M'Kinlay 's extensive and accurate information in regard to
the history and antiquities of his native country are now in a great
measure embodied in Mr. Reid's detailed and excellent History of
the County of Bute.

150 GEOLOGY OF ARRAN.

or Molassus, is well known to Irish antiquaries; it was
printed in the Bollandists' collection, Acta Sanctorum. He
was born in Ireland A.D. 566, educated in Bute by his uncle,
St Blaan, returned for some years to Ireland, and after-
wards, when yet only twenty years of age, retired to an
island in Scotland, where for some years he led the life of a
hermit — probably in the care which is still pointed out as
his in Lamlash Isle — i.e., * Isle-a-Molass,' * later named the
' Holy Isle.' About the year 614 he was elected Abbot of
Leighlin, in Ireland, and was afterwards made a bishop and
apostolic legate to the church in Ireland. He died in the
year 640. The inscription on the roof, in Runic letters, has
no reference to Molassus. The words, ' Nicolas hann raisti
— ' Nicolas this engraved ' — are Norse or Icelandic, and
clearly refer to a Norwegian hermit who resided here at the
time when the Northmen ruled the Western Isles, or about
A.D. 1100. Wilson (Arch, of Scott., p. 531) identifies this
hermit with a bishop of Man. He would make the inscrip-
tion refer to the excavation of the rock. But this has clearly
been the work of the sea; and, besides, the cave was the
abode of St. Molassus at an earlier period. Mr. Wilson
gives some other fragmentary Runes. A small cell or mon-
astery was erected in connection with the hermitage, ap-
parently by Reginald de Insulis, between 1206 and 1212,
probably in connection with the monastery of Saddell, in
Cantire, founded by him, and to which he granted lands in
Arran. It seems to have been an abbot of this small mon-
astery, whose tombstone, bearing his chalice and pastoral
staff, but without any inscription, is still extant in the ruins
of the ancient burying-place and chapel at Clachan Glen, and
which is popularly called St. Molios' grave." The traces of
this small cell or monastery, which was in ruins when visited

* Lasrian and Molassus are radically the same. "Las," meaning
"light," was probably the proper name; the prefix "mo" and affix
"rian" mean "very" and "good," expressing approval or endear-
ment.

EXCURSION X.

151

in 1596 by Munro, Dean of the Isles, were till lately to be
seen north of the cave, marked by an old hawthorn, and
beside a bury ing-ground, which was long used as the chief
place of sepulture for Arran, till the loss of life by the
upsetting of a boat. A pure spring of water close by the
cave — the saint's well — was long famous for its supposed
healing qualities.

Returning to Lamlash, we may examine the ruins of the
old parish church of Kilbride, on the north-east side of the
village, which, though without any architectural beauty, has
some monuments of interest, but of which nothing is known.
The summit level of the road to Brodick is marked by four
granite blocks enclosing a circular space, used as a place of
ancient sepulture, to which allusion will be made in the
sequel, in connection with an account of the Tormore Circles.
A few yards south of it there is an upright stone of coarse
conglomerate, about four feet high, and others near it now
prostrate, all apparently portions of a former circle.

Fig. 29.

152
EXCURSION XI.

TO WIXDMILL HILL AND PLOVERFIELD.

71. The "Windmill Hill is conspicuous from all parts of
the Brodick coast, as the high narrow crest of a long ridge
which divides Glen Shirag from Glen Cloy. It may be
readily ascended from any side. If we pass along its north
front from the String road, we shall see the pits from which
shell limestone, like that at Brodick church, was formerly
quarried ; but the ascent along the south-eastern front will
best expose to our view the structure of the hill. "We pass
up the lovely banks of the Cloy burn, fragrant all the way
with blossoms of the choicest wild-flowers, till we reach the
bridge leading to the old mansion of Kilmichael. This is
the seat of the Fullarton family, proprietors of the estate of
Kilmichael, adjoining, and of that of Whitefarland, on the
north-west coast, the only portions of the island not pos-
sessed by the house of Hamilton. It is not exactly known
in what way the family first came to possess lands in Arrau ;
but there seems little doubt that they acted as stewards in
the island in the time of Bruce, and that the lands of Kil-
michael were granted them in his time; for, in the charter
granted by King Robert III., in 1391, and which is still
extant, bestowing the lands of Ergwhonnyne or Strath-
whillan, Fergus Fullarton is styled " of Kilmichael." The
property of Whitefarland was acquired by marriage. The
name appears to have been at one time MacLewis or Mac-
Lewy, now only preserved in the name of the glen in which
Kilmichael house is situated. The office of coroner of Arran
was hereditary in the family, which is one of the oldest in
the county, and seems to be of the same origin as the
Fullartons of that Ilk in Ayrshire.*

* The history of the family, so far as known, is given by Mr. Reid in
bis History of Bute.

EXCURSION XI. 153

The front of Windmill Hill is formed, through about half
its height, of columnar felspar porphyry, similar to that of
Dun-Dhu, already described (Art. 41). The columns are
four, five, and six-sided, with flat joinings; the pillars lean
in various directions. The junction with the sandstone
below is nowhere visible. About the middle of the front of
the hill a mass of altered sandstone, 12 feet wide, is imbedded
in the porphyry. Close to this is a whin dike, running
N. 10° E. Several others traverse the ridge of porphyry —
another example of the posteriority of the common trap or
whinstone to all the rocks of Arran. The plane of contact
between the sandstone and porphyry gradually ascends west-
ward, and, on reaching the S.W. shoulder of the hill, we
find a wedge-shaped mass of sandstone connected with the
body of this rock below, and apparently separating the
porphyry from the granite which immediately succeeds. No
true contact of these rocks is observable. Sandstone, slaty
pieces of porphyry, granite sand, and bits of granite are seen
lying about, mixed up confusedly, and no line of demarcation
can be laid down. Presently the hill-side shews granite
only. Eastwards all the high ridge of Windmill Hill is
porphyry; sandstone is seen at the north base, but no contact
is visible. The base of the porphyry is an intimate mixture
of felspar and quartz, of a gray or bluish-gray colour, with
imbedded felspar crystals, bits of quartz, and occasionally well-
formed crystals of this substance. We have already (Art. 11)
described at sufficient length the relations of the sandstone to
the Ploverfield granite, to this porphyry, and the syenite of
the hill-sides westwards, into which the granite seems to
shade off, and need not now recapitulate. The series forms
an interesting study. The granite, porphyry, and syenite
are all posterior to the sandstone ; but we have as yet no
means of knowing the age of the porphyry, in relation to the
granite. The common greenstone of the dikes and detached
knolls upon the plateau southwards, is the newest of all
these igneous products. The occurrence in this sharp-crested

154 GEOLOGY OF ARRAN.

hill of an erupted rock like porphyry must be explained as we
explain the existence of granite in mountain peaks, or basaltic
lavas in the mural precipices of Mull and the Giant's Causeway.
72. Passing round the heads of Glen Dhu and Glen Cloy
(Art. 11), we come over the edge of high sandstone cliffs
bounding Glen Cloy on the south. On a grassy ledge under
the western part of these cliffs there occur two remarkable
dikes, producing a highly interesting change upon the sand-
stone. A large body of this rock, between the two dikes, is
altered to the state almost of quartz rock, and beautiful
crystals of amethyst are developed in it; quartz crystals,
both colourless and with a slight tinge of yellow,
also occur. The dike on the S.E. side is a brick-red
porphyry, resembling that on the Corriegills shore ; it appears
also in the cliffs above, where, by wearing, a fissure is
formed upon it; the dike on the N.W. side is of greenstone.
There is a third dike or mass of trap outside the porphyi-y;
but neither its relation to the others, nor of these to another,
can be well made out. The locality altogether is fully as
interesting, on account of the striking metamorphism, as any
in the island. The isolated high " fairy knolls," called the
Sheeans, over the head of Glen Cloy, are trap. Passing
from them to the head of Lamlash, or Ben-leister Glen, we
meet with a few low knolls of like composition, but no body
of overlying trap. Lamlash Glen affords a fine section of
the carboniferous series : red limestone and red marl at the
top of the glen, and lower down two other considerable beds
of limestone, the last being about a mile and a-half from
Lamlash. They contain the usual fossils, and occur amid
massive sandstones. A short distance below the lowest
limestone a dike of felspar rock, or quartziferous porphyry,
fifteen to twenty feet wide, crosses the bed of the river,
ranging nearly N. and S. The sandstone is greatly altered
by it. Between this point and the alluvial plain there are
several veins or beds of claystone and greenstone, breaking
through the sandstone of the river bed.

EXCURSION XI. 155

We can easily pass from this point into the lower part
of Moneadh-Mhor glen. At the opening of this glen, in
the bed and banks of the stream, and of the lead of water
connected with the large mill adjoining, there are two great
beds of pitchstone. They are finely exposed, and exhibit
very strikingly those transitional appearances already alluded
to as marking the relations of this rock to hornstone and
claystone.

Associated with the pitchstones there are claystones, horn-
stones, quartz rock, and porcellanite. Hornstone — that is,
chalcedonic or jaspery chert — seems to be the link between
pitchstone and claystone. Hornstone and pitchstone are both
almost entirely siliceous; the difference consisting in the
colour and degree of toughness arising from a slight change
in composition, or variation in the rate at which they cooled,
or from both. By this change pitchstone passes into horn-
stone. In this hornstone a great many light-coloured spots
with dark centres are gradually developed ; and bands of this
variety succeed the common hornstone. Next to this there
follows a hard quartzose claystone, and the series terminates
in the common claystone of open texture, like that of the
Corriegills shore, already described. The spots or specks are
minute spheres, most probably of felspar or quartz, or of
both, and have probably originated in the manner suggested
at the end of Art. 39. They present a close analogy with
the spherulitic claystone of Corriegills ; but the radiated
structure of the latter does not exist here. Even the larger
spherules of the pisolitic hornstones do not exhibit this
structure. The quartz-rock at the upper end of the higher
pitchstone vein is probably only an altered sandstone. The
porcellanite alluded to above is a white substance, varying
from a dull earthy aspect to that of a white enamel. It
occurs of considerable thickness along the outer surface of
the pitchstone, and is clearly due to a decomposition of this
rock. The incipient stages of this decomposition shew a
structure in the rock which otherwise we should not have

156

GEOLOGY OF ARRAN.

suspected. The surface is traversed by a series of wavy
lines, conforming to one another throughout. This indicates
a laminar arrangement within; and it is along these lines
that the rock splits completely up in the advanced stages of
decomposition.

The relations of these various igneous products is further
illustrated by another dike, which occurs a little way up this
river. A great vein of claystone crossing the bed of the
stream shews a distinct passage into jaspery hornstone. The
jasper and chert veins in the Tormore pitchstone, with
associated claystones and basalts, place these same relations
in a very clear light; and when we view the appearances
which they exhibit in connection with those now described,
we cannot hesitate to admit that all these products of fire
blend into one another, the varying aspects which they
assume being due to a slight change in chemical com-
position, in molecular arrangement of parts, or in the rate
at which they were consolidated from a state of igneous
fusion.

Fig. 30.

The microscopic structure of the Arran pitchstones has
been recently examined by Zirkel of Leipzig (Zeitschrift der
Deutscken GeoL Gesdl 23rd Band., Jan. 1871) and by Mr. S.

EXCURSION XI. 157

Allport of Birmingham (Geol. Mat/., Jan., 1872). Having
submitted thin slices to high magnifying powers, a singular
internal arrangement was brought to light. In the accom-
panying woodcut (fig. 29) we reproduce one of Zirkel's
figures. Mr. Allport's are still more elaborate and beauti-
ful. Describing the section or slice of a specimen from
the Corriegills shore, Mr. Allport says, — "It is seen to
consist of an amorphous glassy base, containing many long
slender prisms of a green pyroxenic mineral, which are
occasionally isolated, but generally form the axes, to which
are attached innumerable minute pale green crystals, arranged
in exquisitely beautiful groups. Some are wonderfully like
fronds of ferns; others have the closest resemblance, both in
form and colour, to microscopic fresh- water algae; in fact,
the field of view seems to be crowded with minute
ferns. . . . The glassy base has a pale yellowish tint
in the open spaces between the groups; but under a higher
power the colouring matter is resolved into a mass of trans-
lucent granules and minute crystals, the latter being much
smaller than the acicular prisms which form the fern-like
groups. A comparison of many specimens, affording grada-
tions in size, shews that all these crystalline particles consist
of the same pyroxenic mineral as the larger prisms which
form the axes. . . . The aggregations of acicular prisms
are invariably surrounded by a border of clear colourless
glass. . . . This glass has not the property of double
refraction; but this power is possessed by the pyroxenic
acicular prisms, which hence exhibit colours. These prisms
are about one-thousandth of an inch long and one-ten-
thousandth broad."

Similar dendritic forms, and others of a bizarre character,
were observed in sections of all the pitchstones examined,
and even in felspar crystals and imperfect crystals of quartz.
They have a striking resemblance to the dendritic forms in
the crystals of snow and ice, which are, in fact, formed
under like conditions of solidification from a state of free

158 GEOLOGY OF ARRAN.

intermobility, under the influence of electrical and other
polar forces. The "pyroxenic mineral," of which the acicular
prisms consist, is afterwards shewn to be augite, and not
hornblende — an interesting confirmation of the view that this
mineral is produced when rocks cool rapidly from fusion, and
become glassy, while hornblende is the result of slow cooling,
when a stony substance is formed, as has been, already
remarked. These papers contain also many interesting par-
ticulars regarding the internal structure of the felspar and
quartz included in the pitchstones and porphyries; but we
cannot follow the subject farther.

In alluding to our account of the hornstones associated
with the pitchstones and claystones in Arran, Mr. All-
port expresses the opinion that hornstone is simply an
altered pitchstone; indicating no doubt thereby an altera-
tion of incipient decomposition, and not a metamorphic
action as by a trap dike. Hornstoue is rather a meta-
morphic claystone, just as quartzite is a metamorphic sand-
stone, and it is thus a link or transition state between pitch-
stone and claystone. But, in fact, the three rocks occur
together, both here and at Tormore, under conditions which
shew that they were of cotemporaueous origin, and that
different rates of cooling or exposure to different agencies
of decomposition have produced the different external char-
acters. This is borne out by the microscopic observations
of Herr Zirkel and Mr. Allport. A specimen of " hornstone"
exhibited " under the microscope the characteristic structure
of the unaltered rock. The quartz crystals [imbedded] are
the same in every respect, but the felspar is no longer trans-
parent, and the clear green acicular crystals have lost their
colour, and appear as yellowish-brown specks."

159

EXCURSION XII.

TO CEJM-NA-CAILLIACH AND THE CASTLES.

73. EXCURSIONS in Arran may be varied indefinitely
according to the taste and objects of the tourist or student.
We have indicated a few best fitted to shew the most strik-
ing geological features of the island; and as we have now
described the different formations and the most remarkable
appearances which they exhibit, we shall only mention
shortly the chief remaining objects of interest in the northern
division, which may be visited in two excursions. We
must then carry the student on three excursions — two by
the South End, and one to King's Cove and the Tor more
Stone Circles.

The principal object in our present excursion will be to
look for "the black crystals," as the smoke-quartz crystals
found in the granite are called. Their chief repository is
the north side of the Suidhi- Fergus ridge and that of the
Castles. They are found in the coarse-grained and rapidly
disintegrating granite of the great northern ridge. A guide
should, if possible, be secured; let him carry a pick-axe to
oper up the disintegrating rock; it will also be well to carry
a pretty heavy hammer; considerable blocks will then be
easily broken. The crystals may also be found by searching
the debris, where each little stream, bursting out of the rock,
forms a talus on the dispersing of the water. Occasionally
beautiful specimens of granite may be found with the in-
gredients crystallized.

The following analysis of the felspar of the coarse granite
was kindly made for me by Mr. Magnus M. Tait, in the
Laboratory of the College, Glasgow : —

160 GEOLOGY OF ARRAN.

Silica, 63'70

Alumina, 20*02

Potash, 12-33

Soda, 1-71

Oxide of Iron, 1'28

Lime, 0'89

Magnesia, trace.

99-93

This analysis shews that the felspar is an orthoclase or
potash-felspar; and that it consists of one equivalent of
tersilicate of potash and three equivalents of tersilicate of
alumina; its formula is KOSSiO2 + 3 (al2 O3 3Si02).

We pass up by the N.E. angle of Glen Sannox, and then
diagonally along the back of the ridge of Suidhe- Fergus.
Here the whole length of the lofty ridge on the south side of
Glen Sannox is before us. It terminates to the left in three
peaks, to one of which, in advance of the others, the name
of Cioch-na-h'oighe is assigned — height 2168 feet ; the heights
of the others are 2541 feet and 2687 feet, that to the south-
west being the highest. The long jagged ridge ends towards
the right in the two lofty summits of North and South
Goatfell.

Ceim-na-Cailliach, or the Carlins' Step, is an immense
chasm or gash in the ridge, overlooked by granite walls
several hundred feet in height. The interior of the fissure
can be easily reached by entering laterally at a pretty low
level on the noi-th side. We find it to be merely a whin
dike worn down to this great depth below the containing
granite. The rock is a dark- coloured, fine-grained green-
stone of loose texture; it exhibits the concentric spheroidal
structure so often alluded to as characteristic of common
whinstone. There is no trace of any other rock; and we
cannot understand how it has come to be so often said that
this dike is pitchstone. The dike and chasm pass down into
Glen Sannox ; but it is very unsafe, if possible, to descend by
this way. The view is very wild and grand. In order to

EXCURSION XII. 161

ascend to Castle Abhael, we must pass north down the
chasm, and then ascend towards the west. Here the glen
and ridges are seen in a new aspect. "When the morning
mists are hovering around Goatfell, and rolling into the
depths of Glen Sannox, now hiding and now revealing the
mountain with the ridges and peaks adjoining, the scene
from the summit of Castle Abhael is extremely grand. —
Various ways are open to us by which to return. \Ve were
once surprised here by a thick fog with heavy rain ; and
guided partly by a compass, and partly by keeping as close
as possible to the junction of the coarse and fine granite, we
found our way safely along the edge of the great circular
ridge which runs from the north-west base of Castle Abhael,
in the direction of Glen Dhu, at the upper part of North
Sannox. But the convenience of our companion, with his
load of " black crystals," must now be consulted in regard to
our homeward route.

162

EXCURSION XIII.

GLEN SANNOX AND GLEN KOSA.

74. THIS is a favourite excursion with visitors to Arran ;
many will devote a day to it in preference to others which
are less easily taken from Brodick as a centre. But there
are comparatively few objects of geological interest, and we
shall not therefore enter into any detailed description.

There is an interesting old cemetery at the entrance of
Glen Sannox. It is all that remains of a chapel dependent
on Kilbride church, and dedicated to St. Michael. A rude
image of the saint is to be seen upon a stone built into the
outside of the wall. The house was probably connected with
the Abbey of Kilwinning, to which Sir John Monteith, Lord
of Arran, granted, in 1357, the lands of Sannox and patron-
age of the churches. The barytes mill, which till lately
marred the solitary grandeur of the scene which opens as we
reach the plateau at the mouth of the glen, was entirely re-
moved several years ago by order of the late Duke of Hamil-
ton. It was erected here to grind the sulphate of barytes,
or heavy spar, raised from veins which traverse the old red
sandstone. These are seen in the bed of the stream, running
in a direction nearly N.E. and S.W., and dipping variously;
they appear also on the hill-side southward. The junction in
this burn is not well seen, but interesting junctions occur
under the base of Cioch-na-h'oighe. By the bank of this fine
stream of crystal water, rushing over its bed of granite sheets
and granite sand, amid huge rolled masses of the rock, is a
delightful walk. Near the head of the glen we may diverge
a little to the right, and examine, at this lower level, the
dike and great chasm descending from Ceim-na-Cailliach.
Following it southward on its line of bearing we trace it

EXCURSION XIII. 163

entering the granite precipices in front of the base of Cior-
Mhor, and passing on over the col, in the direction of the
axis of Glen Rosa. The easiest pass into this glen is at the
western side of the col or ridge joining Cior-Mhor to the
base of Goatfell. In the hollow up which we pass we have
a whin dike beneath our feet, and granite walls on either
hand — a pathway, in fact, has been formed by nature in the
disintegration of this dike. On reaching the summit we
observe a great dike, most probably a continuation of this
one, ranging right up the front of Cior-Mhor. When one
reflects upon the mode of origin of these two rocks, granite
and greenstone, it strikes one with wonder to perceive the
curious relations which they maintain, and the important
part in the physical condition of the region which the dikes
play. If the molten matter of greenstone had had full vent
here and overflowed, the interior pressure would have been
relieved, and the mountains of the granite nucleus would
have stood at lesser altitudes. M. Necker goes so far as to
suppose it quite possible that the excavation of Glen Rosa
may have been determined by the great dike above noticed.
In the lower part of the glen the river runs upon it, between
high granite walls, for a long distance ; and it crosses S. into
Glen Shirag. It is thus by far the most continuous dike yet
traced in Arran.

Up till the year 1822 this path was known as practicable
only by shepherds, some of whom occasionally used it. But
in that year two enterprising young ladies, Miss Alison and
Miss Crooks, both from near Kilmarnock, but residing at
Brodick, having arrived on a summer afternoon at the top of
the ridge by passing up Glen Rosa, determined to try the
descent into Glen Sannox, and return by the coast road.
With great difficulty and loss of time they made good the
descent ; but were so late on arriving at Brodick that all the
young men of the village had started off in parties, in differ-
ent directions, to search for them. Their tale excited no
small wonder.

164 GEOLOGY OF ARRAN.

If the climber so desire, he can easily pass into Glen lorsa
from the head of Glen Sannox, across the col between The
Castles and Cior-Mhor; he will find some good plants by the
way, and from the top of the col reach the summit either of
Cior-Mhor or of The Castles without difficulty.

A very interesting dike exists on the ridge between Glen
Sannox and Glen Rosa. It intersects the high southern part
of the ridge between the two glens, near the point where
North Goatfell starts up from the level of the col. The dike
here consists of green, or nearly black, pitchstone only ; but
the sides are not seen in contact with the adjoining granite,
nor is the dike traceable downwards into Glen Sannox. It
ranges, however, nearly due west across the head of Glen
Rosa, a little below the col, and thence right up the face of
Cior-Mhor, where it is found to consist of a central band of
green pitchstone, with six other bands, three on either sida
The first of these, next the pitchstone, is a quartzose claystone
approaching to hornstone, with light-coloured spots in the
base, apparently both of quartz and felspar ; to this succeeds
a band of hard claystone porphyry, outside of which on
either side is a broad band of fine-grained greenstone with
imbedded crystals, constituting the variety called trap
porphyry. The width of the vein in the first part of its
course cannot be determined; in the front of Cior-Mhor it
varies from 20 to 30 feet. The vein forms a ravine up
which one can clamber to the summit. Above this ravine
the claystone disappears, and the pitchstone central band
widens from 2 or 3 to 5 feet, while each of the trap bands is
about 4 feet. As illustrative of these curious relations, the
remarks in Art. 72 may be referred to.

75. There is little else of much interest in Glen Rosa ; the
granite junction has been described already in our First
Excursion (Art. 26), and the singular chasm just noticed will
be seen as we pass along. The adder (Pdias bents) shown
in the annexed wood-cut (fig. 30) is often met with
in the glen, on the dry parts of the path, and in places

EXCURSION XIII. 165

where there is a dry bottom under the heather ; and
the tourist in crossing these will do well to use caution.
Again and again we have narrowly escaped treading upon
them in such situations here, on the North Sannox shore,
in Lamlash Glen, and other places. We have never seen
them higher, however, than the top of Glen Shant Rock,
1100 feet ; nor are they as abundant on the west as on the
east side of the island, on account, no doubt, of the shelter
and warmth. They are found also in Holy Isle. The ocur-
rence of these creatures thus on islands is very singular, yet
part only of a great physical problem, like the existence at
the present day of the tiger and elephant in Sumatra, or the
former existence of the hysna, hippopotamus, elephant, and
other wild beasts in Great Britain ; or the camel-leopard in
Greece and Malta. Man does not often interfere for the
dissemination of such creatures ; yet a case of such interference
has occurred in our own time. The late Mr. Cleland, of Ban-
gor, Downshire, a naturalist of some note, attempted to intro-
duce into Ireland the creature of which we are now speaking,
as well as the toad. Several individuals of both species wore
placed in his "policies," but in a few months one after another
was found dead. It is very unlikely that such benevolent
intentions were entertained by any naturalist of old towards
the inhabitants of Arran or the Isle-a-Molass. The size of the
adder in Arran seldom exceeds 2 feet — a common size is 18
inches ; but we know a case in which a size of 2 feet 5 inches
was reached. The adder is the only British representative of
the poisonous group of serpents; but there is no authenti-
cated case of fatal effects resulting from its bite in Arran or
elsewhere in Britain. We have heard of more than one
case of severe symptoms ; but these were removed in a few
days by an application of herbs boiled in butter. On one
occasion only have I noticed aggressive attitudes in the
adder. I was searching for ferns among the crevices of the
limestone and sandstone beds on the north front of Maoldon,
when a strange, prolonged h-i-s-s startled me; but I had no

16G GEOLOGY OF ARRAN.

idea whence the sound came, or how it was produced ; and it
was a good many seconds before I caught sight of a large
adder, within less than a foot of my hand, in an attitude as
if to spring, with erect crest, open mouth, and tongue
alarmingly visible —

" Attollentem iras et coernla colla tumentem."

On throwing in a small stone — for I have never felt in-
clined to kill an adder, having already a preserved specimen
— the creature retired into its hole.

The Gaelic name of this glen is said to mean the Ferry-
point Glen. This name " Ferry-point," as indicating that
water may have filled the glen since man inhabited the
island, is remarkable, especially when taken in connection
with the anchor found in the glen. This was discovered, not
on the hill-side, where peat is now cut, as stated in Art. 25,
but on the level of the river above the parks, where peat was
formerly cut. Art. 25 was printed off before the error of
our informant was known to us ; but it is clear that on
evidence so slight as this, we cannot assign to a very great
antiquity the use of iron anchors, nor a modern origin to
the old sea-cliff which surrounds Arran > for it could only
have been when the sea stood at that old level that Glen Rosa
was occupied by the sea. That remarkable changes have
taken place in Brodick Bay in late times we know. The
Duchess Anne built of Arran oak, and in order to present it
to the government, a vessel of considerable size, to the west
of the group of ash trees near the present manse, and launched
it there. And later still, about 120 years ago, it was usual
to ship cattle, at high water, in the Cloy, from a natural pier
of stone a little way west of the present smithy garden, north
of the road.* But such changes as these of the recent period
are of quite a different character from the occupancy of Glen

* I am indebted to the kindness of Miss Brown of Invercloy for this
and many other interesting facts regarding local traditions and
antiquities.

EXCURSION XIIT. 167

Rosa by water: floods, tides, and storms may have produced
the former — the whole island must have been raised 25 feet
at least to lay dry the " Ferry-point " Glen. This name, it
is true, may have reference to a lake ; and certainly the
present mound is very like the remains of an old lake barrier.

The .burns of Arran ai'e all beautiful and picturesque, but
in different styles, according to the nature of the rocks across
which their courses lie. Those of the granite and slate tract
are perhaps not the most beautiful ; but this Rosa burn,
especially in its lower course, and after it enters Glen Shant,
surpasses in romantic beauty most of the bright streams of
this lovely island.

76. Over the wood which covers the eastern end of the
ridge on the south side of Glen Rosa a curious meteorological
phenomenon is often witnessed. When the wind is at S.W.
or W., with a damp and warm atmosphere, a column of
vapour is seen ascending from near the centre of the wood,
remaining in a nearly steadfast position for some time, and
then suddenly vanishing; to be again formed, and again as
suddenly disappear ; and this is repeated through a period
sometimes of several hours. For some time we supposed
that it was the smoke from a gipsy fire : but the explanation,
no doubt, is, that currents of wind descending Glen Shirag on
one side, and Glen Rosa on the other, produce, by the rapid
out-draught into Glen Shant, a partial vacuum and vortex of
light ascending air over the middle of the wood, which is
completely sheltered ; and by this relief of pressure a con-
densation of vapour takes place, just as a cloud is formed in
the exhausted receiver of an air-pump.

We may mention now, when on the subject of meteorology,
an interesting case illustrating the relations in which the dif-
ferent kinds of clouds stand to one another : — On a bright hot
day, after rain on the day previous, a pretty rapid current of
wind from W.S.W., with scattered masses of white fleecy
clouds, prevailed aloft from early morning as a land wind;
while a light east wind blew below as a sea breeze. The cur-

168 GEOLOGY OF ARRAX.

rents met, a little way out in the channel, off the Corriegills
shore; and along a strip of sky, stretching from north to
south, the fleecy cumuli of the S.W. current were arrested by
the current from the east, and resolved into a broad band of
cirri. By four o'clock in the afternoon the upper current had
descended to the surface ; and a strong breeze from the west,
preceded by a whirlwind, immediately sprang up. Towards
evening this gradually subsided into a gentle air. The day
was one of those delightful Sabbaths to be enjoyed only in
the rural districts of Scotland or the North of Ireland, The
Free Church congregation of Brodick was observing the
communion services; and a vast concourse of pious and most
attentive worshippers was assembled from all parts of the
island ; one minister officiated in English in the church,
while another conducted the service in Gaelic from a tent iu
the open field. Apart altogether from the solemn feelings
awakened by the simple and touching service, the mere
picturesque adjuncts of these scenes at Brodick will never
fade from the memory of any one who has witnessed them.

The Adder— Fellas berus.

169

EXCURSION XIV.

BY THE EASTERN SHORE TO DIPPEN AND LAG.

77. WE enter now upon a region whose geological struc-
ture is more simple, and strata less disturbed. The rocks we
shall meet with are sandstone, felstone, and trap. The
sandstone is rarely conglomerate, and has a greater admix-
ture of argillaceous matter. Having been deposited as a fine
sediment in quiet waters, and afterwards consolidated by the
pressure of superincumbent ocean, aided by heat below, it
was affected by a general movement and extensively frac-
tured. Through the fissures thus formed streams of melted
matter were forced up from lakes of molten rock in " the
nether depths," and overflowed from hundreds of vents the
surface of the sandstone. Cooling under varying local condi-
tions, and having a slight local variation in its constituent
parts, this matter became felstone in one part and trap in
another, and differs from lava consolidated in the same way
simply in this, that it cooled, not in the atmosphere, as lava
does, but under a great pressure of incumbent ocean. Now,
it must be remembered that the granite and slates of the
north plunge beneath these sandstones, and must also have
been pierced by the streams of liquid rock. This escape of
molten trap, on a large scale, from beneath the granite,
relieved that rock and the whole southern district from an
enormous upward pressure; and hence the granite was not
raised in a solid form ; and hence, too, the contrasts between
the northern and southern divisions of the island. Liquid
matter poured out from beneath could not consolidate in the
form of steep ridge or beetling precipice; the disruptions
attendant on the elevation of the land, and the action of
currents, would impress upon the surface most of the

170 GEOLOGY OP ARRAN.

existing forms ; ice-action and ordinary disintegration would
do the rest.

Sandstone is seen all round the shores of Lamlash Bay.
This bay, in fact — the best harbour on the Clyde — has been
excavated wholly in sandstone; and such is the nature of the
rock, that the bottom affords admirable holding ground, while
the water is deep enough to float the largest ship. Holy Isle
forms a breakwater, with safe and deep entrance at either side.
" What stupendous might was exercised," says the Rev. Dr.
Landsborough, " when this gigantic mole was pushed up
through rock and earth and water, and the elevated sandstone
overflowed by a stream of melted porphyry ! Behold the
power and goodness of God ! How many, after weathering
the storm, and casting anchor under shelter of this mighty
breakwater, have exclaimed : ' Thanks be to God, we are in
Lamlash Bay,' " * The shores exhibit in a striking manner
the wearing action of the sea at a later period; dark trap
veins intersect the shores in all directions, and some of these
between Clachland point and Lamlash stand up like huge
ramparts above the beach, shewing the extent to which the
softer rock has been worn.

78. The distance from Lamlash to Lag by the mountain
road, crossing the Ross and descending Glen Scorodale, is ten
miles; by Dippen, twelve miles; or if we diverge to visit Kil-
donan Castle and the southern cliffs, fourteen miles. From
Lamlash to Whiting Bay the distance is about four miles. As
the path by the shore has not much of new interest, we take
the public road, which affords finer views of the scenery, and
then pass down to King's Cross, at the S.E. angle of the bay.
Here rocky ledges rise from a pretty pebbly beach, from
which Bruce is said to have embarked when entering on his
]>erilous enterprise of rousing the men of Carrick. A cross,
no longer to be seen, commemorated the event ; and close
adjoining are the remains of a small fort, erected apparently

* Excursions in Arran, chap. ix.

EXCURSION XIV. 171

for the purpose of defending this landing-place, one of • the
best in the bay and at its entrance. There is a considerable
body of coarse greenstone here intruded as a bed among the
strata of sandstone, pieces of which are entangled in the
greenstone, and altered to the condition of quartz rock. The
greenstone is traversed by long continuous veins of a finer
kind approaching to basalt, probably cotemporaneous. The
rock is divided into large rhombic blocks, but the surface has
no resemblance to a basaltic pavement. On the shore at
Whiting Bay there are multitudes of dikes, appearing in
singular relations to one another — intersecting, bifurcating,
uniting, and separating again, entangling and altering the
sandstone, and so on — relations that it would be tedious to
describe in detail. The shore is covered with granite
boulders, one of which, near the south-east end of the bay, is
larger than any boulder south of Brodick, with the exception
only of the Corriegills boulder. It is remarkable that one so
large should be found so far from the granite centre.

79. A little farther S. a fine glen opens to the west; this
is Glen Eais-dale, at the head of which there is the highest
waterfall in the island, bearing the name of Eais-a-Chranaig.
The stream descends over trap and sandstone in two fine
falls through more than 200 feet ; and here, as throughout
the rest of its course, the scenery is highly picturesque. On
the south side there is a trap dike, 10 feet wide, cutting the
sandstone and passing into the overlying mass of greenstone.
The contact of these overlying masses with the sandstone
shews a remarkable change. The rock has lost its red colour,
and becomes a white felstone, slightly porphyritic. Farther
down, the ordinary laminated character is gradually assumed.
On a terraced bank to the right of the fall there are the
remains of an ancient fortress, concealed amid wild wood,
which appears to have been the strongest of all the forts of
the ancient inhabitants, with the exception of that at Druma-
doon. The wall, formed of huge slabs, without lime, is 25
feet thick and 90 yards in circuit. " From the situation of

172

GEOLOGY OF ARBAN.

this strength, and the vast labour which has been employed
in rendering it impregnable to the attacks of an enemy, it is
probable that it was used as an encampment by the early
islanders for the security of their families in the case of
invasion."* The local antiquarians point out the "hill of
council " adjoining, where the chief men met, in case of alarm
by beacon fires on the heights of Dippen or Dun-fion ; and tell
you that there was a larger population in old times than now
along these glens and sheltered slopes. On the brow of the
terrace over the waterfall there are several slabs of greenstone
marked with glacial striae, directed nearly east and west, that
is, in the direction down the glen. The hills westward are
all of overlying trap, and in some places shew veins of pitch-
stone. This terrace, the lowest of several ascending steps,

sweeps round south-eastwards in a high cliff, and comes upon
the shore at Leargie-beg, where for more than a mile south-
wards it forms sheer precipices of highly prismatic rock resting
on sandstone (fig. 32), which it covers for a long way with its

* M' Arthur's Anti'juities of Arran, p. 90.

EXCURSION XIV. 173

talus of fallen blocks — a very perfect counterpart, though on
a small scale, of the great basaltic fagades of Mull and the
Antrim coast. Inland, the ground rises in successive ter-
races formed by the sheets of greenstone laid one over the
other like the steps on a pyramid. Under the cliffs the
scenery is wild and romantic, and the place altogether
affords the best field in Arran for the study of the basaltic
rocks. The peculiarities of soil and shelter have favoured
the establishment of a few uncommon species of plants. On
the level of the beach S. of the talus we first come on a
great bed of claystone, very like the great Corriegills bed,
hard, splintery, and breaking into small prisms. It is seen
in the sandstone slopes above, but does not intersect the
columnar facade of greenstone overlying the sandstone — this
fixes the relative age of these two erupted rocks. The sand-
stone now first appears on the shore, and is immediately
traversed by a trap dike four yards wide, ranging N. 10° W.,
and this, as well as the claystone, produces a decided alteration
on the structure of the sandstone. But there is nothing on
this part of the coast that equals in interest a pavement of
basalt which extends for 200 to 300 yards in breadth, and
from the sea line to the base of the cliffs. This causeway
forms the nearest approach we have anywhere seen to the
basaltic pavements of Staffa and the Giant's Causeway. The
rock is not all a pure basalt; much of it has a fine porphyritic
structure. The pillars are less perfect, less various in form,
and are without the cup and ball articulation, the test of a
perfect pillar; and many cracks crossing the surface somewhat
mar the uniformity of the prismatic structure. This latter
defect is perhaps owing to the action of the tide, which covers
the greater part of the bed; a portion of which, indeed, seems
to have been already removed by this cause. As this great
lava stream rushed through the sandstone strata, pieces were
torn off, borne along, permanently entangled in it, and are now
found isolated in the basalt and altered to a crystalline
structure. But there was another later irruption — the

174

GEOLOGY OF ARRAN.

basaltic bed is traversed by a greenstone dike five yards wide,
and ranging W.N.W. This, as well as other dikes, traverse
to the very summit of the cliffs, intersecting the overlying
columnar greenstone, which is thus older than the greenstone
of the dikes, but newer than the claystone already noticed as
clearly erupted. We have thus here erupted rocks of three
ages.

80. The sandstone, gradually ascending southwards, now
rises to the edge of the cliff, and the overlying trap retires
inland, but continues to occupy higher terraces over the sand-
stone. In front, near the cottages, the shore line is low, and
from this point southwards the trap appears in many interest-
ing relations with the sandstone, to which we can only
invite the observer's attention, without attempting to describe
them. The trap is an injected bed, and occurs in some
places overlying, in others in repeated alternation with, the
sandstone, and again the two rocks come together at the
same level. The dikes, too, are very singular; they make
strange turns and bifurcations, and one consists of alternate

Fig. 33.

bands of trap and sandstone. Presently after, a low cliff
begins and runs westwards by Kildonan Castle. This inter-
esting old keep (fig. 33) stands on the front of a flat terrace,
which runs back northwards to the trap hills, and faces the

EXCURSION XIV. 175

sea in a cliff fourteen or fifteen feet high. Its foundations are
laid on a rock of columnar clinkstone with much iron, like
that of Holy Isle, and which changes inland to what is con-
veniently but incorrectly called a trap porphyry, a gray or
blue compound base with earthy felspar imbedded.

81. Kildouan Castle is a square keep, without ornament,
four storeys in height, with several vaulted apartments. The
date of its erection is uncertain ; but it was probably put up
at the time of the war with the Edwards, as one of a line of
watchtowers reaching from Ailsa Craig to Dumbarton Rock.
Like that of Looh Ranza, it was a royal castle till 1405,
when the lands and castle were granted to Stewart of
Ardgowan, in whose family they remained for 150 years.
To attach a turbulent chief of Cantire and the Isles —
Macdonald of Dunivaig — the Eai'l of Arran, then Regent,
forced a sale upon the Stewarts; but a few years afterwards
he bestowed on Macdonald, in exchange, valuable lands at
Saddell in Cantire, and became himself owner of Kildonan.
The Stewarts, of two families, seem next to have got the
property; first, one under Cromwell, the Hamilton estates
having been forfeited ; and again, another by succession.
They went afterwards to the Earls of Bute, from whom they
were purchased, about thirty years ago, by the Hamilton
family. The late Duke, when Marquis of Douglas, erected
the handsome shooting-lodge which now stands on the sum-
mit of the Dippen cliffs, with a magnificent look-out over
rock and flood. Kildonan Inn is by the shore, under the
platform, where we may rest for a little and partake of its
hospitality.

82. The claystone "platform on which Kildonan Castle
stands extends inland as a bed in the sandstone, which also
shews below it on the shore, and forms the basement of the
terraces and ridges of trap which sweep round westwards by
Benan and Lag. This trap is mixed up with claystones, and
dark-based felspar and pearl-stone porphyries, in such a way
that it is impossible to unravel the order among these pro-

176 GEOLOGY OF ARE AX.

ducts, if order there be. Sandstones, shales, and marls
occupy the shore, and stretch far up the river-courses and
lower parts of the glens, whose higher sides exhibit in section
the series of traps and felstones just alluded to. It would
be desirable to have all the strata thoroughly examined for
fossils, as these, if found, would fix the age of the beds.
These interior sections are more deserving of inspection than
the shore, which shews only a repetition of appearances with
which we are familiar — dikes of greenstone, claystone, and
rarely pitchstone, some of the dikes standing up like walls,
and numerous granite boulders of moderate size. There are
in all about forty dikes, and the range is generally a little
"VV. of N. One of the largest is that near Kildonan, forming
the little harbour of Drum-la-borra. A dike attaches Pladda
islet to the mainland, and has suffered so little disintegration

*

in this exposed situation that at high water a vessel can cross
only at one place.

Auchinhew streamlet comes down opposite to Pladda, and
shews some fine sections, and a very singular and highly
romantic waterfall and pool called Eais-a-Mhor, or the great
fall. The fall is N.E. of the Auchinhew farm-house, N. of
the road; but the section lower down should first be visited.
There are here several alternations of columnar trap and
amorphous trap, with sandstone and marl. The horizontal
joints and vertical seams are both very distinct, and mark off
most of the pillars as pentagonal prisms. At the fall the
water descends through a height of seventy or eighty feet into
a large circular basin, or amphitheatre, the perpendicular
sides of which present a facade of pillars resting on
sandstone. Over this is another bed of sandstone, sur-
mounted by beds of amorphous trap, which forms the channel
of the burn above, towards the moors, and by its hardness
has prevented disintegration farther up. The front of the
fall is formed by a dike, which, when it reaches the basaltic
stratum, divides into two branches and terminates, thus
clearly shewing that it formed the vent of eruption. The

EXCURSION XIV. 177

sandstone is indurated along the lines of contact, and the
strata shifted out of parallelism. The effects of so small a
stream are quite wonderful to contemplate here, and suggest
the idea that it must have been of larger volume in bygone
times. Ballymenooh stream, a little way east, shews some-
what similar appearances; and farther west, near Benan-
head, on the north side of the road, there is a fine section of
the same kind. The stream has cut down the strata to a
depth of seventy feet, in a fine cascade; and a trap dike,
after traversing the sandstone strata, loses itself in the over-
lying greenstone — another clear case of a lava-chimney.
Calcareous nodules, rarely exceeding six inches in diameter,
are disseminated through the sandstone, both here and in
several other sections; and Mr. Headrick mentions (p. 117)
that a small cascade on a rocky front south of Auchinhew
burn, has deposited a mound or small hill of stalagmites, or
lime incrustations, shewing that lime prevails in all the
sandstones here, and hence that these probably are the true
carboniferous beds. The high grounds north of Benan-head,
which reach 523 feet in altitude, consist of felspar porphyry,
and greenstone, the former generally in a superior position,
but in some places coming against the latter at the same
level.

83. The Struey rocks and Benan-head form, both for the
geologist and botanist, one of the most interesting features of
this coast. The headland is 457 feet high, and is a mass,
from top to bottom, of igneous products, variously inter-
mingled, but consisting chiefly of a dark-based felspar por-
phyry, basalt and greenstone. The crystals imbedded in the
dark compact felspar base are of that variety of felspar called
pearlstone, with which are commingled quartz crystals of a
smoky hue. In the greenstone there are dark-coloured
crystals of hypersthene with the lustre of crystals of oxide of
iron. It is difficult to make out exactly what are the
relations of these rocks to one another in the headland. In
some places they are intermixed, and perhaps alternate in

178 GEOLOGY OF ARRAN.

bands ; but the porphyry, as a mass, may be said to prevail
upwards, the two other rocks being in greater proportion
below, except that on the east side the porphyry comes down
to the level of the beach. The entire mass of the headland
is wedge-shaped, narrowing downwards; the tipper edge of
the porphyry cliff being on the same level on either side as
the strata of sandstone, the ends of which abut against the
trap rock throughout. Though the sides converge rapidly,
the termination of this wedge-shaped mass is not seen, the
igneous rocks running out under low water and cutting off
the sandstone. Dikes of common trap intersect the whole
series from top to bottom of the cliffs ; and a little way west
of the headland a dike of quartziferous porphyry, very
distinct from the porphyry of the headland, intersects the
sandstone 011 the beach.

The Black Cave in the front of the headland has been
excavated in a vein of rotten trap, between two basaltic
walls, and extends about forty paces by fourteen, with a
height of seventy or eighty feet. Rocks have fallen from the
roof making an opening overhead. The vista across the sea
is finely closed by the erupted cone of Ailsa. Loitering here
may turn out to be very embarrassing, as the front of the
cliff is not passable at or near high water; and at certain
times of tide we may get into such an awkward situation
that our only chance of escape is " to swim for it."

The cliffs westwards present for some distance porphyry
and greenstone, apparently in alternating bands; but it is
difficult to make out the exact relations. Thence to the
mouth of Torlin Water, the shore and cliffs are occupied by
variously coloured sandstones and marls, with occasional
nodules of limestone and ironstone. The platform of the
beach is crossed by multitudes of dikes, mostly of basalt and
greenstone, which usually produce a decided alteration on
the sandstone strata. The road from Benan to Lag, which
runs a little way inland, shews generally, along the undulat-
ing grounds on either side, beds of claystone beneath the

EXCURSION XIV. 179

surface soil ; in the hollows in some places greenstones
appear, either the tops of dikes shewing here in their inland
course, or the edges of beds emerging from under the clay-
stone. After our hard day's work, we are glad to find that
Lag is less than three miles distant from Benan-head. At a
turn of the road and fall in the ground near the Kilmorie
post-office, we see the old church of Kilmorie, perched on a
high bank over Torlin Water; and soon the welcome sight of
the peaceful hollow and neat inn greets the eye — we shall
find few inns anywhere in which there is the same attention
to our wants, and the menage so unexceptionable.

180

EXCURSION XV.

LAG TO THE SHELL BEDS.

84. LAG is an admirable centre for several excursions. A
thorough examination of the shell-beds, especially with the
purpose of collecting the leading species, will take two days
at least. There is much to see, very curious and interesting,
by following up to tneir sources the two largest streams,
Torlin and Slaodridh Waters ; Benan-head and Dippen-cliffs
may very well be visited from it, reversing our order; as also
the shore both ways, east to Benan-head, and west to the
porphyry of the Brown Hills. For the Botanist and the Anti-
quary there is a fine field; and to a lover of the "gentle art"
the two largest streams of the island, with their two con-
siderable tributaries, will afford sport for several days. But
"The Lag" has other charms; its situation is peculiar and
romantic; its complete seclusion and perfect quiet wonder-
fully tranquillizing ; though warm and sheltered in its sunny
dell, the inn is not ten minutes' walk from the sea — an early
stroll there braces for the day's work ; the country about is
the best cultivated and richest in Arran; and it is delightful
to saunter at even-tide among the sweet teeming fields, to
mark the fitful changes of the sea, to watch the sun go down
behind the Cantire hills, and catch his last rays gilding the
dome of Ailsa,* "The Lag," as it is called in the neigh-
bourhood— that is, the Hollow — owes its peculiar features
to a speciality in the rock formations : — there has been an
upthrow and fracture of the sandstone-beds, here soft and

* The farms of Clavich (Mr. Robert Speirs) and Torlin (Mr. Fiulay
Cook) are noted for the admirable methods pursued upon them, and
the successful results.

EXCURSION XV. 181

marly, attended by the appearance of great dikes of basalt;
the stream, issuing by a narrow opening and liable to great
floods, would readily disintegrate and sweep off the materials
of these fractured beds. The part most deeply imbayed is
that immediately opposite the entrance of the stream and its
tributary into the existing hollow, below Kilmorie Church ;
from this the sweeping currents would be reflected to the
heights north of the inn, and thence again towards the point
where the stream escapes. The principal dikes fcre at this
point, below the bridge. The largest is thirty to forty feet
wide, ranges nearly N., and forms the high cliff over the
flower-garden; another, farther S., is ten feet broad, and
ranges nearly N.E.; a third, nine to twelve feet wide, ranges
N. 10° E., and dips W. at 80°. Above the bridge there are
two large dikes; the lesser stands high above the south bank
of the stream, like the side of a former dam, and is seen on
the north side in contact with disturbed and altered sand-
stone— width twenty-five to thirty feet, range N. 26° W.,
dip E. at 70°; the greater, farther up the stream, is thirty-
five to forty feet wide, and ranges N. 5° E. The sandstone
strata in the hollow dip usually about S. 10° to 20° W., at
angles varying from 20° to 30°; but there is considerable
disturbance. The heights round the basin vary from fifty to
seventy feet above the stream.

85. The beds of Arctic shells most accessible from Lag,
and the most satisfactory to examine, are about three-
quarters of a mile distant, and on the banks of a tributary
to Torlin Water, called the Cloinid or Clenid burn. It
joins the Torlin on its right or west bank, a little way N.E. of
Lag Inn, coming down from the north-west through a deep
rocky gorge, easily passable when the water is low. Another
way to reach them is to follow the high road as far as a lane
near the Kilmorie school-house, pass by this to the river, over
which there is a wooden foot- way, and then to cross the
fields, by a cottage on the high bank opposite the church,
down into the channel of the Cloinid burn. We are thus

82 GEOLOGY OF ARRAX.

on the shell-beds at once. They rise steeply to a great
height on the east bank of the stream, a very little way to
the north of the upper end of the gorge. Fragments of
shells on the surface soon shew that we are in the right
place. In an excursion to these beds, and the others in the
neighbourhood, running through three days, I was accom-
panied by the Rev. H. W. Crosskey of Glasgow, who has
carefully studied the shelly deposits of the Clyde basin, and
east of Scotland : and the result of our joint observations is
the account that follows : — The object of enlarging the list
of species already given (Art. 21) was quite subordinated to
that of working out the physical order of the beds, and
establishing among them definite relations. It has been
already remarked (Art. 21) that under the terms Boulder-Clay,
or Till, many distinct beds have been classed as one forma-
tion ; whereas there can be no doubt that if the subject were
properly investigated and understood, order would reign
among these as among the other strata of the earth's crust,
and that beds of a like mineral character would have also a
definite stratigraphical position. Not that such a series of
beds is to be supposed to occur universally — for this is not
the case with the solid rock formations themselves; but
that if there be a considerable development of these " super-
ficial beds" in different places, they shall be capable of sub-
division into several terms, with distinct and persistent
characters and definite relations to one another.

Now, in attempting to establish a physical order among
these beds, we are met by two difficulties which we do not
encounter in the case of rocky strata. The upper layers
under the soil are loosely aggregated and pervaded by fissures
which admit water, while the lower layers are retentive of
water. Thus a great hydrostatic pressure comes to be exer-
cised, and as the beds have not that continuity and coherence
which enables one part to lend support to another, the upper
beds slip down upon the lower. Such landslips also occur by
the undermining of a stream. In this way the order of the

EXCURSION XV. 183

beds is completely masked; a bed ten, twenty, thirty feet
higher in the series is placed in front of perhaps the lowest
bed, and so remains for a long period, till by successive slips
or underminings the whole fallen mass is removed, and the
natural order of the beds is again revealed. In most cases
satisfactory evidence that such landslips have taken place
will be had by examining the grassy brows along the summit
of the cliffs. Then again, the beating of rain, the action of
the air and sun, and the trickling of runlets of water over
the front of such a section, produce a sort of general wash-
over, as we may call it, which gives the same facing to all
the beds, and completely obscures the natural succession.
There is a third obstacle and cause of error, like what one
sometimes meets with in the case of the rocky strata — the
difficulty of finding junctions. The wash renders it impos-
sible to see the line of contact, and it is often necessary to
search over a large surface. The observer takes the field
armed with a hammer only, but finds his weapon, good for
other purposes, of no use in this contest. A pickaxe and
spade are required for opening up the beds, and clearing
away loose matter, in order to obtain a distinct view of the
planes of contact.

The difficulty of finding a true section, and so determining
the actual succession of the beds, is now apparent; and the
great probability, that, under the name of boulder-clay, beds
really distinct have been included. And we cannot help
thinking that much of the confusion which prevails has
arisen from overlooking the causes of error now adverted
to. One of the sections under consideration was visited by
us three times, and another twice before we were able to
satisfy ourselves about the order of the beds. In several
places we left the sections completely opened up for the
benefit of future observers : but the elements of change may
soon alter the features.

86. The base of the section in the Cloinid burn is the
true old boulder-clay resting on the upturned edges of the

184 GEOLOGY OF ARRAN.

sandstone rock, which dips nearly S.W. The depth is
about fifteen feet in the lower part, but the upper surface
declines southwards, or down the burn, so that the bed
becomes thicker upwards. The clay is excessively hard
and tough, and the hammer makes no impression upon it ;
the imbedded stones are of all sizes, from small pebbles
to large boulders, all confusedly mixed, and many striated,
smoothed, or polished. It is very striking to observe the
prevalent rounding-off upon the edges of a large proportion
of the stones, and the high polish which has been given to
them, and that upon all alike, of whatever size. This seems
clearly to indicate a powerful agent acting for a long period.
It cannot have been a continued transport in water, for, as
already observed (Art. 22), this action would have left no
trace of striation. The majority are local, but comparatively
few of sandstone; porphyry, greenstone, and syenite, which
form the higher grounds and the hills northwards, abound,
but there is also granite of the coarse-grained or Goatfell
variety, and slate, both, of course, from the northern moun-
tains. The ordinary wash covers, in most places, the surface
of the boulder-clay, and in this wash adhering to the surface,
we found a few small fragments of shells; no fragment in
the clay itself. — Over this is a clay bed, the repository of the
shells, sinking southwards with the upper surface of the bed
below. It is seven to eight feet thick, and is a hard
compact, rather fine and dark clay, with a good many
small stones, not striated as a whole ; our experience, indeed,
is, that few were striated. The deposit is very different
from that below; much less compact, more easily worked,
and the contrast of the stones is remarkable. The upper
part of the bed is of a reddish colour, and a little sandy?
while, in the lower part, the colour has more of a bluish
tinge, like the boulder-clay, shewing that it was formed on
the spot, and partly out of the boulder-clay, on the bottom
where the shells were living. The arctic character of this
bed is clearly shewn by two of the s}>ecies found in it

EXCURSION XV. 185

— the Pecten Islandicus, and Astarte borealis — the only two
absolutely arctic. By far the greater number of shells
are in single valves or in a broken state, yet not
so small but that the species can be determined. This
fragmentary condition of the shells shews a very disturbed
state of the waters along the coast on which they lived.
But even on stormy sea-coasts, with stony shore, sandy
mud, and broken shells, there are sheltered places inside of
points, and behind banks and ledges of rock; and there can
be no doubt that, somewhere about here, if one could
happily alight upon it, there must be some such bed of
unbroken shells indicating a quiet deposit.

The following section gives the succession of beds at this
place in ascending order : —

1. Boulder-clay, twelve to twenty feet thick.

2. Clay-bed with shells, seven to ten feet.

3. Dark sandy bed with open texture, four or five feet —
apparently local.

4. An upper drift of sand and stones; thickness variable.

5. A compact bed of stones, with less sand, forming a
marked line on the cliff, but not easily reached; thickness
five or six feet.

6. An upper drift, similar to No. 4.

7. Surface soil.

These upper drifts, covered in front with their facing, or
wash, look very like the boulder-clay; but a closer examina-
tion, and removal of the false face, shew them to be very
different. They are much less compact, more loose and
easily broken, generally more sandy; and the stones are
mostly angular, or, if polished, the surface has a finish very
inferior to that of the stones in the boulder-clay; as if
they had been exposed, since the polish was put on, to
some continued wearing action. In fact, they have very
much the appearance of a river deposit; and in looking
at some of the heaps of detritus left by floods on the sides
of the stream below, it struck us how very like they were

186 GEOLOGY OF ARRAN.

to these upper drift beds. The young observer could not
employ a day given to geology to more purpose than in
examining these upper drifts and contrasting them with the
boulder-clay, in some typical section.

The shell-bed No. 2 is seen over the boulder-clay, from
the extreme southern end of the section at the steep bank
over the stream, across the whole front of the section,
ascending, however, northwards; and the same species are
got in it throughout, there being a great many fragments of
Pecten Islandicus, Cyprina Islandica, Hodiola modiolus,
Astarte borealis, A. compressa, A. elliptica, and occasionally
single perfect valvea The part of the bed most accessible
for working is on a slip towards the southern part of the
section.

A little way up the stream, on the same side, there is a
very high bank with apparently a like succession of beds,
but so steep that it is very difficult to get a footing ; and, as
the day was now far spent, we were obliged to leave them
unexamined.

87. On the west or right bank of Torlin "Water, opposite
Kilmorie Church, and a little below the wooden foot-way,
the shell-bed is again seen. The sandstone strata here
dip N.E., opposite to the usual direction; the stream for
a short distance runs upon a whin dike, and there are
beds of trap farther down on the opposite bank. The
shell-bed rests on the sandstone, thinning out southwards in
the bank. The section has a small horizontal extent, the
bank northwards being broken and covered with wild
shrubs; but there are indications of the existence of the
boulder-clay in this direction, and it probably comes on here
tinder the shell-bed as the ground rises up the river. Over
the shell-bed there is a red sandy bed twelve to fifteen inches
thick, over which there is an upper drift of earth and stones.
The shell-bed has the same character as in Cloinid burn ;
it is tough and compact, dark coloured, with small stones,
and the same species of shells. Not having previously

EXCURSION XV. 187

compared the two beds in Cloinid burn, an observer would
be very apt to confound this shell-bed with the boulder-clay.
88. We pass now to Slaodridh Water, about a mile west
of Lag. Less than half-a-mile above the bridge, a large
tributary from the north-west joins the Slaodridh. This
is the Crook-crever burn, on whose east bank in several
places the shell-bed occurs. A very little way up the burn
there begins a highly picturesque gorge, with steep banks of
wild wood, cut by the stream through beds of variegated
sandstone, alternately hard and soft, crumbling shale falling
fast away, and the hard sandstone projecting in ledges.
A good way up, this gorge expands into a wide open
space, and high clay banks form the eastern side, but im-
mense slips greatly confuse the order of the beds. After a
repeated and careful survey, and having eliminated the
various sources of error, we made out the succession of the
beds as follows : — Sandstone rock is the base of the whole
section; over this a bed of the true boulder-clay, thirty
or forty feet deep, reaching about half-way up the bank,
the clay being of the usual hard compact, almost unworkable
structure, and with some very fine examples of striated
stones, blocks, as well as others of lesser bulk. Its upper
part becomes, through a space of nearly a foot, a hard,
dark-coloured gravelly sand, extremely compact and ob-
durate under the pick-axe. On the top of this, and strongly
contrasted with it, rests the shell-bed, of almost exactly
the same composition as that already described in the other
two sections; a dark compact clay, with small stones rarely
striated — veiy different from the boulder-clay in the com-
parative facility with which it is worked. In addition to
the shells already named, most of which were found, there
was a perfect Leda, the valves united, and fragments of
Balanus. Above this bed are the upper drifts very similar
to those already noticed, and here, as in the former case,
very strongly contrasted, if carefully examined, with the
boulder-clay; yet, there can be very little doubt that they

188

GEOLOGY OF ARRAN.

have often been confounded. This section shews the series
very distinctly, and it was left by xis completely opened up.

89. The last section opened was just below the east
end of the sandstone gorge, opposite a bend on the river,
and a little way above a wooden guard placed upon the
stream. Here a blue argillaceous shale, dipping S.W. at
15° to 20°, forms the bed of the stream, and rises up
with the same dip into the east bank. Over this the
boulder-clay and other beds come on; and, following ap-
parently the inclination of the shale below, they slope
towards the stream, and so thin out westwards. This ar-
rangement is shewn in the annexed cut (fig. 34) : —

9

Fig. 34.

a a a, Shale ; &, boulder-day ; c, shell-bed ; d, upper drifts.

At the expense of some time and labour, the facings
and debris were here completely cleared away, so that the
entire section was laid bare, from the underlying shale to
the upper drift bed; and nothing could be more interesting
or perfectly satisfactory than to observe the contacts and
complete distinction between the contiguous beds. The
boulder-clay has all its usual characters, as already described,
and does not yield a single shelly fragment ; the shells begin
to abound directly the overlying clay bed is entered. This
has its usual character, which we need not again describe,
and over it come the upper drifts, again strongly contrasting
with it and with the boulder-clay, but here not well ex-
hibited, from the nature of the ground.

90. This section is highly illustrative in another way.

EXCURSION XV. 189

We often find that, either from their successive erosion,
or from some inequality of original deposition, these beds
thin off at the out-crop, just as rocky strata do; and thus
the beds are brought so close together that the distinction is
not perceived by an unpractised observer; and that shells
from an upper bed get on to the surface of a lower, or
perhaps into its very substance. This source of error is
most likely to be met with in the case of basin-shaped or
nearly level deposits. We think it most important that
all who engage in the study of these most perplexing
superficial formations should guard against this source of
error and those already indicated (Art. 85) ; they ought
first to study the mineral character, and then the order
of the beds, and not hastily assign to the boulder-clay — that
is, the lowest and oldest bed — shells or other fossils which
may really belong to those which are superior to it. We
have never found fossils in this oldest bed, while we were the
first to shew that marine shells and elephantine remains
occur under it. We by no means assert that fossils may not
exist in it, yet we cannot help thinking that far greater
caution should be used in assigning such to this bed than
has hitherto been shewn. Ita upper surface undulates, and
the shell — bearing hollows of the bed above might readily be
taken as belonging to this.* We cannot theorize with any
certainty respecting the origin of these deposits until, by
a more careful study and extended examination, we get
hold of a number of great general facts in regard to the
physical order of the beds, their internal structure, and
fossil contents. In regard to the fauna of the Arran beds, it
will have been observed that it is all of one character ; the
same species occur in all the beds, with slight local excep-
tions. The shell-bed itself is of uniform character, with
very slight variations; and hence it follows that uniform

* Let every student before entering on these beds peruse carefully a
most lucid and instructive paper by the Rev. H. W. Crosskey, on
their structure and relations (Tr. Phil. Soc., Glasgow Sess. 1868-69).

190 GEOLOGY OP ARRAN.

conditions affecting the marine creatures prevailed over this
district. But the area is too limited to warrant any general
conclusion. The probable conditions have been already
noticed (Arts. 20-22). The boulder-clay may have been
formed by laud ice; the shell-bed over it, under and around
a rim of ice when the land had been depressed; and the
upper drift beds partly by local glaciers as the ice was
disappearing, and partly perhaps by rivers, then of larger
volume than now, as already hinted. The elevation of the
shell-bed, or of the upper surface of the boulder-clay (Art.
21, last sentence, in which this idea is obscurely expressed),
taken in connection with the depth due to the species,
gives us the measure of the depression of the land. Now,
the elevation of the beds in question ranges from 70 to
180 feet; so that the greatest depression of which we have
any evidence here is 180 feet below the present sea level, not
taking account of the depth required for certain species ; and
we hesitate to speak with confidence of any greater depres-
sion, with our present knowledge of the beds farther up the
glens. In Clachan Glen there are beds of at least equal
altitude and enormous thickness ; but their altitude has not
yet been determined.

91. The bed and banks of these four streams present many
objects of interest besides the shell-beds. They exhibit sand-
stones and shales with nodular limestone, and sometimes
nodular ironstone, among which no fossils have yet been
found, and these rocks are pervaded by traps and felstones
of all varieties, both in dikes and overlying masses. Near the
head of Crook-crever burn claystone appears in both relations :
on the bank it overlies, in the bed of the stream it breaks
through the sandstone in conformable beds, and produces a
very decided change upon it, altering it to the condition of a
fine white quartzite, a very clear proof that the claystones in
many places are not altered sandstones, but themselves
erupted rocks. The bed of the Slaodridh at Glen Rie mill,
and the country thence to Lag, and again far up the stream

EXCURSION XV. 191

on the east side, shew felstones as the prevailing rock — on
the west side from Glen Rie upwards the prevailing rock is
greenstone. At Glen Rie mill the sandstone is greatly dis-
turbed and altered by a prismatic compact felspar, with bits
of quartz and rarely felspar, which forms a bed in the sand-
stone.— We might return from Lag this way to Lamlash,
and trace the carboniferous strata across the watershed of
the Ross. We might pass also up the banks of the Torlin,
and trace out a wondrous variety of curious relations among
the traps and felstones which abound there, visit a Stone
Circle not far from Urie Loch, examine the syenites and
pitchstone which break out on Urie Hill, and then pass down
to Whiting Bay or Lamlash. An account of the rocks to be
seen on both these routes would lead to great detail, and
imply much repetition; but the walk of ten or eleven miles
will be found very instructive.

On a bank near the sea, within the first field east of the
mouth of Torlin Water, there is a fine specimen of the
sepulchral cairn. It consists of a great number of stone
chambers, formed by five large slabs, said to be filled with
human bones. One only has been opened, and from it
certainly bones and a skull were removed. The lid of the
stone cist or chamber still lies near; an upright stone pro-
bably marks the head. It was once surrounded by a wall ;
but though now unprotected it will not be further disturbed
without express authority. Nothing is known regarding its
history. A little to the west of the mouth of the Slaodridh,
on a high, steep mound near the shore at Haddock Port,
there are the remains of a fort called Tor-chastel, or the
castle hill. It is connected to the mainland behind by a
narrow neck, which is guarded by an outwork with a stone
foundation. On the top there is the circular foundation of
a Danish fort, consisting of a wall of large stones without
cement, five feet thick, and having a circuit of about fifty-
two yards. It was doubtless made strong to defend this
place of landing for enemies from Cantire, Haddock Port

192 GEOLOGY OF ARRAN.

being the extreme south-western point of the island, and the
only port here. Bones of man, the boar, deer, and ox have
been found among the ruins; and the surface has been in
later times cultivated, as ridges are marked on the summit.
Not far from this point is the natural harbour of South End
between the mouths of Slaodridh and Torlin Waters. This
is formed by four dikes, of which two run nearly N. and S.,
and two nearly E. and W. ; of the two latter, the seaward
one is highly prismatic, and stands up like a wall, but is
breached, and admits vessels, while it forms a shelter for
them. The other forms a rude landing pier on the north
side.

The old parish church, called St. Mary's cell, or Kilmorie,
existed early in the fourteenth century, about the time of
Bruce. It was however a mere cell, being only nineteen
feet by ten. The present church, built in 1785, is near the
site of the old chapel, and is said to have the font which
belonged to it.

193

EXCURSION XVI.

TO DRUMADOOX AXD KING'S COVE.

92. To King's Cove is a favourite excursion, and deservedly
so, for the scenery there and about Drumadoon is extremely
beautiful. With this may be combined a visit to the new-
granite tract of Craig-Dhu, the wonderful assemblage of
pitchstone veins, and splendid group of stone circles, at Tor-
more. In Arran as in Switzerland, the success of a day's
work depends upon an early start. But we must be sure of
the day; let it be calm and sunny, if the scenery on the
charming Drumadoon shore is to be enjoyed to the full.
Leaving Invercloy, we turn to the left, pass Brodick Church,
and begin the long ascent of the String road. On the right
we look down into the " Sleepy Hollow" of Glen Shirag, and
note with surprise what work of excavation the little streams
have done at their last joyous leap into its depths. On our
left the ground swells rapidly up from Brodick wood, and
culminates in the high porphyry ridge of Windmill HilL
From behind the smooth ridge of Old Red in the foreground,
the granite peaks and crests rise up in succession as we ascend,
and appear in new and grand aspects. As we breast the
slope, we turn often to gaze in wonder and delight on the
lovely scene we have left behind — the smiling plain of
Brodick, the glassy bay, and hanging woods. At the summit
we are five miles on our way, and have climbed 600 feet ;
before us is a fine vista, shut in by two long ridges, and
closed by Kilbrannan Sound and the hills of Cantire.

After the watershed is passed, granite appears on the left
hand — the north-west limits of the Ploverfield district — and
runs a long way down the side of the hill ; then bands of

o

194 GEOLOGY OF ARRAX.

syenites, the same as those which succeed the granite on its
west border (Art. 11), strike in upon the road, intermingled
with sandstone, here of carboniferous age. The bottom of
the glen and steep ridge on the right are composed of the
old red sandstone. The old bridle path runs at the base of
this ridge. On its smooth outline it is curious to notice
watercourses and glens in an absolutely initial stage; such
is the force due to the height of descent, that water
collected into the merest rill, by the slightest possible
inclination, has a great power of excavation, and speedily
forms a perceptible channel and considerable talus. Leaving
the high road at the Glen Loig bridge we shall find a bed
of limestone high on the southern hill-side, by following a
bridle road leading up to an old quarry. It is a crystalline
limestone without fossils, much altered by trap veins which
traverse it, and to which the rock adheres firmly. It
seems to be near the base of the carboniferous series, in
the same position as the limestone at Achab farm and the
Fallen Rocks.

"We are now close upon the Craig-Dhu granite, already
fully described (Art. 12). Craig-Dhu is a high steep cliff,
fronting that part of the Shiskin road where the Dougrie
road parts off from it. A talus of blocks strews its base (see
Art. 12).

A little farther forward, opposite the pillar letter-box, the
Dougrie road turns off to the right, crossing Mauchrie Water
by a substantial bridge, and passing through the stunted
remains of an old forest; the distance is about three miles.
The only house on this road is the farm-house of Mauchrie,
opposite to which a path leads down to a wooden footway
across the stream and near the shore by which we can easily
reach the Toruiore pitchstones. The stone circles, a mile
inland, and due S. of the farm-house, may also be reached
this way by a very short route. But we shall visit them
from the shore.

From the wayside post-office we continue our course to

EXCURSION XVI.

195

Black water-foot, a distance from Tnvercloy of eleven miles,
passing on the left, the glens of Craig-Dhu, Balmichael, and
the Clachan, and having a wild moor on the right, on whose
northern brow the tall standing-stones of the "Druid Circles"
aiTest our gaze. At two miles distance we pass Shedok
Inn, and two miles farther, turning sharp to the right, we
are at Black water-foot, the extremity of the vale of Shis-
kin. The unpretending little inn here will afford most
neat and comfortable quarters should we find occasion to
remain over night.

93. On the south-west coast, in both directions from this
point, there occurs the greatest development in the island of
the felstone family of igneous rocks — a peculiarity for which

Fig. 35.
a, Sandstone ; b, broken band with cave below ; c c, claystone.

it does not seem possible to assign any cause. At Black
water-foot a great body of indurated prismatic claystone
occupies the bed of the stream, the sides of the little port, the
shore for some distance, and a long line of inland cliff. Its
outgoings N. are lost amid the sands. In some places it
forms very perfect columns, with a near approach to the true
cup and ball jointing. In the cliff it intersects the sand-
stone in the form of great dikes; and in one place is seen to
turn up the eiids of the strata abutting against it ; in another

196 GEOLOGY OF ARRAN.

place the beds approach it at an inclination, and there is a
mass of broken rock between them, a small cave being
formed in this at the bottom. Here, as shewn in the annexed
cut (fig. 35), the mass of claystone on the south side is
divided into great concentric curved bands, arranged round
a prismatic nucleus, and exhibiting on a large scale that
peculiar structure of the rock, often seen in the fracture of
hand-specimens, which it has in common with the other
felstones and the ordinary traps, and which comes most
strongly out in decomposition. The appearances here are
strongly illustrative of the eruptive character of the clay-
stone.

The sands stretching hence to Drumadoon exhibit many
dikes, ranging in various directions, of various widths, and
consisting of basalts, greenstones, both common and of
porphyritic structure, and felstones. The botany of the
sands is rich; the surface has a thin carpeting with a pro-
fusion of beautiful wild flowers. On approaching the grand
facade of columns forming the precipice of Drumadoon, we
find at its south-west base on the shore a great bed, or per-
haps dike, of felspar porphyry tilting up the sandstone, at a
high angle, in no way connected, so far as can be seen, witli
the overlying porphyry above, but running out into the tide-
way, and forming a low cliff in the sands. This is Druma-
doon Point. It is intersected by basaltic dikes, one of
which has been hollowed out by the waves, and forms a
landing port for boats. It is probable that these beds, and
others on the south-east side, are connected, underneath the
sandstone, with the overlying masses. The beds below can-
not have been, at a former period, continuous with the
porphyry of the facade, as the sandstone rises high in front,
between them. Yet the cliff has obviously suffered consider-
able disintegration along the front; an isolated, nearly entire,
pillar stands out from the facade at the south-west corner,
and the sandstone below is strewed with great numbers of
fallen columns, forming a talus to the precipice. These

EXCURSION XVI.

197

features are well shewn in the cut annexed (fig. 36), which
represents the cliffs at Kingscove, with Drumadoon Point
in the distance.

The porphyry of the cliff is in columnar forms, the
pentagon being the prevailing figure, and there are irregular
flat jointings; the pillars are from 1 to 2 feet in the side,
and the height 80 or 100 feet at the middle or highest part
of the precipice; the entire height being about 230 feet.
The porphyry has a base of compact felspar or clinkstone of
a whitish or brownish gray or dark-blue colour, and im-
bedded crystals of glassy felspar or pearlstone, common

Fig. 36.

felspar, or earthy felspar, in a state of decomposition, and
with an unctuous feel, like steatite ; there are also transparent
bits of quartz, easily known from felspar by not shewing any
laminated structure, with the form and lustre of crystals
whose angles have been worn off; the base is often dark and
has cavities lined with black oxide of iron. The rock is
thus a quartziferous felspar porphyry, often ironshot. The

198 GEOLOGY OF ARRAN.

rock thus closely agrees, in mineral character, with the
porphyry of Dun-fion (Art. 42). Hand- specimens, shewing
only the earthy imbedded felspar, have the look of amygda-
loid ; and there is a variety in which the felspar crystals
have decomposed out and been replaced by calcareous spar
by subsequent infiltration ; and another in which quartz
becomes a component part of the base, and the felspar
crystals disappearing, the rock has a granitic structure, like
that of Eurite (Art. 50). In some specimens the imbedded
felspar has a nucleus of quartz. The pillars rest on a
laminar stratum, consisting of the base of the porphyry
without the felspar crystals, and thus resembling a dark
claystone, metamorphic sandstone, or some varieties of the
lighter coloured basalts. A thin stratum of white sandstone
is interposed between this bed and the underlying red sand-
stone. There are two dikes traversing the cliff right up
through the columns of porphyry. There is a path by the
base of the precipice, which is about 500 yards in extent ;
and off its N.W. termination there are masses of a similar
porphyry on the shore and in the low cliff, whose relations
to that of the great facade cannot be made out.

Dminadoon, the hill with the fort — Drum or Drim,
whence the Latin Dorsum, is a rounded ridge — is so named
from a fortress on the summit, which seems from its position
and remains to have been one of the strongest places in Arran
in the olden time. A wall eight or ten feet thick, founded
on huge blocks, starting at either end from the front of the
cliff, sweeps round landwards so as to enclose several. acres.
Mr. Headrick, writing in 1807, says — "In the middle of the
arch there is a gateway, on each side of which are great
heaps of stones, which seem to have been additional works
for its defence; within the enclosure are several ruins of
houses of loose stone. . . . Had it a sufficient supply of
water, it might be rendered impregnable." Many of these
stones have since been used in the erection of fences and
cottages, and there are no traces of any ruins. The massive

EXCURSION XVI. 199

blocks which still remain are too numerous to have been all
found upon the summit in the manner usual with boulders,
and must have been dragged up the eastern slope by some
powerful mechanical appliances. The great area and the
form shew that its purpose was rather a place of refuge, in case
of danger for the families and cattle of the neighbourhood,
than a fort to be defended by warriors. In later times it
became by universal consent a sanctuary or place of refuge.
On the farm of Drumadoon, S.E. from this fort, there is
a fine Cromlech — an immense flagstone resting horizontally
on upright stones deeply sunk in the earth {New Stat. Acct.
Hute, p. 54, 1837, 1845). Tradition has fixed upon it the
name of " The Grave of Fingal's Daughter." It was opened,
and found to contain an urn with ashes. There were at
one time many remains in this part of the island : a huge
cairn at Black water-foot, a tximulus in Shedok farm, on
which there is now a clump of trees ; the " Panel's Stone,"
where Fingal is said to have held his court of justice ; St.
Columba's seat, a cairn on which the saint is believed to have
rested when passing through the country on his mission to
the heathen natives ; a few isolated monoliths and other
remains, besides the great Tormore circles, to be presently
described.

94. The coast now sweeps round in a fine bay, whose
grassy floor and low cliff of red and white sandstone contrast
finely with the rugged features of Drumadoon. Northwards
among these sandstones there are inclined beds of green
pitchstone, which alter the sandstone strata in contact ; their
termination cannot be traced. There are also bands of
claystone in the cliffs. Rounding the angle of the bay, we
come on high cliffs of soft yellow sandstone, deeply excavated
by the sea, when it stood at the higher level so often already
alluded to — now and for more than 2000 years above that
ever reached by the highest tides. To one of these great
interest has been given by the tradition, that it was for a
short time the residence of King Robert Bruce, on his land-

200 GEOLOGY OP ARRAN.

ing from Eathlin, in 1307, before the seizure of Brodick Castle
and descent upon Ayrshire, though there appears a greater
probability that Loch Ranza was the port where his flotilla
found shelter. A hard vein of sandstone had stood in the
cliff vertically, and the strata of softer rock declined either
way, from some early fracture; on this the sea working for
ages scooped out the cave. The hard vein is still seen in the
sharp Gothic-like arch above, in the position of a keystone,
and in the back part of the cave comes down to the floor,
forming a column with a recess on either side, 30 feet in
length. The cave is 100 feet long, 50 wide, and 55 high.
The column has figures rudely cut, representing a two-
handed sword and a cross. There are also rudely-drawn
plans representing a scene in the chase, and also sheep,
goats, and cattle. These are all assigned by tradition to the
time of Bruce's sojourn in the cave. The figures are men-
tioned by Pennant, writing in 1772. " I felt a holy
veneration," says Mr. Headrick, "while exploring the cavern
where Bruce had sheltered; not because he was a king, but
because he was a patriot, a hero, and an assertor of the
independence of his country ; " and then pictures his
struggles and aspirations while working here his great plans ;
and ends by an eloquent denunciation of all warriors who
have fought for other objects than those which Bruce
pursued.*

The great hero of the Gael, Fioun or Fingal, is said also
to have resided here, and to have had a son born to him in
the cave ; and a straight groove is shewn on the side of the
cave, more than two feet long, said to be an impression of
the child's foot the day after his birth; from which Mr.
Headrick, with great gravity, goes on to calculate that
Fingal must have been seventy or eighty feet in height, and
his wife fifty or sixty !

There are several other caves near the King's Cove; a
large one on the south side, with two entrances and a huge
* Hcadrick's Arran, p. 1GU.

EXCURSION XVI. 201

pillar between, is "the King's stable;" smaller caves on the
north are " the King's kitchen, larder," and so on.

95. The King's hill, u Tor-an-righ," a very little way
east of this, is the highest point on the shore here, being
350 feet in height ; and has probably given the name Tor-
more, or the great hill, to the townland. On its summit
there are many granite boulders of the fine variety, among
which one was distinctly marked with glacial strise. The
cliff here is the side of an immense dike of felspar por-
phyry ranging nearly N. and S., and fronting the sea
in a mural precipice sixty or seventy feet high, along the
base of which there is no passage except near low water.
The porphyry is very like that of Drumadoon, and the
dike is eighty or ninety feet wide. Placed longitudinally
in this dike, westward of its middle part, there is a
greenstone dike about four feet wide, running out with it
seaward to the low point where the porphyry dips into the
water; it is also crossed diagonally by a greenstone dike,
which bifurcates upon it, and is seen far up on the hill-side
above. Alongside the dike there is a large cave. The finest
set of dikes to be seen in Arran occurs here, exhibiting, in a
small space, all the members of the igneous series, greenstone
and basalt, porphyry and trap porphyry, pitchstone, clay-
stone, and hornstone. To understand the relations of the
various dikes, the shore should first be traversed several
times, the whole extent being only a few hundred yards.
The dikes range over the sandstone platform to the north-
west of this. The principal dike, traceable continuously for
a long distance, is formed of green pitchstone; it rises from
the sea southward with a range of N. 40° E.; a width of ten
or twelve feet, and a S.E. inclination of nearly 30°; but the
course undulates 35° or 45°, bending towards the west, or
into parallelism with the shore, towards its northern termina-
tion. There are veins of slaty hornstone on both sides,
next the sandstone, and on one side a thin layer of basalt.
On the side of the great vein next the sea, numerous veins

202 GEOLOGY OF ARRAN.

enter it nearly at right angles. First, a vein of hornstone
six to eight inches wide; next, one of basalt five feet wide.
Another vein running a little N. of W. has on its north side
three feet of pitchstone, passing into hornstone, then four to
six inches of claystone, and on the south side fifteen feet of
basalt or fine greenstone. Next is one thirty feet wide
running N. 35° W., consisting of basalt four feet, clay-
stone fifteen feet, basalt again six feet, containing a pitch-
stone vein along its middle part; lastly, a vein of greenstone
five feet wide. There is here also a vein of trap porphyry
four feet wide ; this and the others intersect the pitchstone
vein, but the intersection is obscured by debris. The last vein
northwards is one of pitchstone with hornstone and jaspery
quartzite, running oblique to the shore, and varying from
eleven or twelve to thirty feet. Just outside the tideway it
meets and enters the great pitchstone vein, which, in its sweep
north, has reached this point, ranging now N. 10° E., and
at the junction there is nothing visible but claystone, which
forms the flooring at the common point of union. The two
veins unite here; and the joint vein continues its course
under masses of sandstone, but is not seen in the cliff aloft,
nor can it exist there. Still farther north, near where the
cliff ends, we noticed a pitchstone vein running E.N.E.,
seven to ten feet wide, and visible in the cliff above; and
this appears to be the same vein continued. The close
association of so many igneous rocks is extremely curious;
we may either suppose the various products to be cotempor-
aneous, so that different rates of cooling and varying chemical
composition may have produced the differences now observ-
able ; or we may ascribe them to successive eruptions taking
place under varying conditions. Professor Jamieson was
the first to describe this interesting locality, which he does
with great fulness and accuracy. He gives also a figure
illustrating the position of the secondary or cross veins (Min.
of Scot. Id., vol. i. pp. 17 and 102).

The following analyses of the claystone and pitchstone

EXCURSION XVI.

have been made by Mr. Magnus M. Tait, F.C.S., formerly
assistant to Dr. Anderson, Professor of Chemistry, Glasgow
University, now of Messrs. Vivian's Works, Swansea : —

Claystone.

Silica,
Alumina, .
Potash, .
Soda,
Magnesia,
Oxide of iron,
Lime,
Water, .
Loss,

Pitchstone.

Silica,

Alumina,

Soda,

Potash, . . .
Water and organic matter,
Magnesia, ....
Lime, .....
Oxide of iron, ....
Loss, .....

72-50
11-53
5-24
3-37
2-72
2-06
1-79
0-70
0-09

100-00

66-03
12-55
5-02
4-13
4-20
2-33
2-80
2-75
0-19

100-00

96. A walk in the late afternoon from Black water-foot to
Lag, across Leac-a-Bhreac, forms a pleasant close to the day's
excursion to King's Cove. Leac-a-Bhreac, or the Brown
Hill, is a high, round ridge, with steep seaward front, and
long inland extension, lying between Kilpatrick, near Black
water-foot, and the mouth of Slaodridh Water. It is com-
posed of felspar porphyry, being the largest mass of this rock
in the island. The base is a gray, or bluish gray, compact
felspar, sometimes passing into a jaspideous hornstone, often
iron-shot, and the imbedded crystals, glassy felspar; smoky
quartz also occurs in round bits, like worn crystals, and the
rock affects the columnar form, the columns having flat

204 GEOLOGY OF ARRAN.

joints; but the form most frequent is that of rhombic blocks.
The surface decomposes into a dark brown layer, from the
atmospheric action on the iron of the base. The hills get
their name from this circumstance. The rock is well adapted
for ornamental purposes, and could easily be procured in any
quantity. The porphyry occupies the shore, cutting off the
sandstone, and forms considerable cliffs, which exhibit the
usual wearing action of the sea, when at its old level. The
caves are a shelter for the sea-spleenwort and other good
ferns; and one of the largest is often used as a preaching
station in connection with Kilmorie Church. The " grave?
sweet melody" of the Scottish psalmody here mingles finely
with the mellowed sound of the dashing waves. On the
S.E. side, towards Slaodridh Water, north of the road, the
porphyry is seen to rest on sandstone, which, rising gradually
from below it, begins to occupy the lower grounds seaward,
the porphyry retiring inland, and ranging across the moor to
the Crook-crever burn, where it is associated with claystones.
Farther up the Slaodridh Water, there are again found great
beds of a similar porphyry, as already described in Excursion
XV. The great development of these felstones is a remark-
able feature of the south-western portion of the island.

Leaving Tormore, or King's Hill, and crossing to the
Shiskin road, we are in front of the fine-grained granite tract
of Glen lorsa, and accordingly we find that here the great
proportion of the boulders are of this variety of granite ;
while those on the eastern side, and along the southern
plateau, are of the coarse variety. This points to a cause
locally acting in the direction of the glens : we never find
lorsa granite on the east side of the island, both kinds are
mingled on the west side (Art. 22).

If we return to Invercloy on foot, the walk will be varied
by passing up Clachan Glen and down upon Glen Cloy. We
strike off the main road about a quarter of a mile north-east
of Shedok Inn, at the bridge over the Clachan burn, and first
examine the curious old effigy in the centre of the small

EXCURSION XVI. 205

cemetery, popularly called St. Molios' Grave, but in reality,
we believe, the tombstone of an abbot of Saddell, with his
chalice and pastoral staff (Art. 70). The place of worship
adjoining is a chapel-of-ease to Kilmorie Church, near Lag,
and has service every third Sabbath; on the two other
Sabbaths the hearers attend the Free Church on the west
side of the vale, most of them having sittings in that church
also — an example of liberal Christian feeling which it would
be well to follow. The detrital accumulations at the mouth
of Clachan Glen have been already referred to (Art. 90).
They consist of the true boulder clay below, with striated
stones, and above of the upper drifts, the intermediate shell
bed not being present in any of the sections, so far as we
could discover; but our examination was not lengthened^ and
the difficulties are not less formidable than those to be met
with in the sections already described (Art. 85). The lower
and middle parts of this glen shew the lower portions of the
carboniferous series ; and high on the south side, apparently
by an upthrow of the strata, beds of white limestone, with
quartz pebbles, like that of Achab farm and the Fallen
Rocks, are brought to the surface. The metamorphic action
throughout, but especially towards the head of the glen, is
the most extensive and remarkable we have seen in Arran —
beds, bands, and dikes of claystone, greenstone, basalt, and
syenite, intersect the sandstones and marls, or are intercalated
with them, and produce such marked changes that all dis-
tinction between sedimentary and igneous products is
obliterated. The sandstones pass across the watershed into
Lamlash Glen on one side, and Glen Cloy on the other.
From the high plateau above Ploverfield, we may descend
either by Windmill Hill or the Sheeans. To walk from
Invercloy over the " String " by Mauchrie water-foot and
Tormore, and return by this route, is no mean performance
for one day. But the platform ought to be gained before
sunset, as twilight falls rapidly in Glen Cloy, and the ground
between the Sheeans and Invercloy is broken and difficult.

206

ANCIENT REMAINS IN ARKAN.

97. PREHISTORIC and other ancient remains are not so
numerous in Arran as its extent would lead us to expect.
The physical character of the surface limited, at all times as
now, the residence of its people to a narrow belt of low land
round the shores, and offered few temptations for the access of
fresh settlers ; while its remoteness from the mainland served
as a security against invasion or sudden surprise, and so
rendered unnecessary the erection of strongholds to defend
the coast. When by the advent of peaceful times the inhabit-
ants began to partake of the general improvement of the
country, and, by mixing with strangers, to lose that super-
stitious regard to old monuments which once prevailed, in
reclaiming waste places, monuments which stood in the way
were without scruple removed. Many perished thus in the
end of the last century and beginning of the present, which
the improved sentiment of the present age would have care-
fully guarded. Happily some time since this work of
thoughtless destruction was arrested in Arran by the in-
structions mainly of the late Duke of Hamilton, a man of
taste and culture, and lover of art and of the antique ; and we
may safely reckon that by the judicious and watchful care of
the present respected agent for the family in Arran, Mr.
Paterson, of Whitehouse, Lamlash, the instructions referred
to will be acted upon over the whole of the island.

98. The ancient remains in Arran consist of cairns, barrows,
or mounds, monoliths and stone circles, chapels, cells, forts,
camps, and castles* Examples of all of these still remain;
but we know, from the early accounts of Arran which we

ANCIENT REMAINS IN ARRAN. 207

possess, that many have been entirely obliterated or removed
— several of those still existing have been noticed by us
already in our various Excursions ; others which we may have
overlooked, or which no longer exist, we shall now notice,
and then refer to our own share in endeavouring to illus-
trate the structure and uses of the most remarkable Arran
monuments yet remaining. — Pennant (Tour in Scotland,
1772) mentions that there are in Glen Cloy "five earthen
tumuli or barrows placed in a row, with another on the
outside of them. On the top of one is a depression or hollow,
on that of another is a circle of stones whose ends just
appear above the earth. These are probably the memorial
of some battle ; the common men were placed beneath, the
plain barrows, the leaders under those distinguished by the
stones. A mile farther is a retreat of the ancient inhabit-
ants, called Torr-an-Shiain Castle [fairy's mound], sur-
rounded with a great stone dike. Here Robert the Bruce
sheltered himself for some time under the protection of
Mac-Louis." This is noticed in the new Statistical Ac-
count, 1837 (published 1845), as "the vestige of an ancient
fort in Glen Cloy, where Bruce's partizans who arrived
in Arran before him sheltered themselves, while the English
held Brodick Castle." In the same work (" Buteshire,"
p. 23) imperfect remains of stone circles are mentioned as
existing at the top of Blairmore Glen, at the head of Glen
Cloy, and of several sepulchral cairns on Blairmore farm, near
the manse of Kilbride. "But the largest cairn which the writer
(Rev. Dr. Allan M'^Taughton, minister of Kilbride, 1837)
has examined is one of more than 200 feet in circuit, at the
head of Moneadh-More Glen, "W. of Lamlash. Its position
was near the mouth of a narrow pass between the E. and W.
sides of the island. Most of the stones were carried away
two years ago to build a stone fence in the neighbourhood
and several stone coffins, each composed of six unhewn flags,
were found under them. Similar coffins are every year met
with in cutting drains and ditches in different parts of the

208 ANCIENT REMAINS IN ARRAN.

parish, sometimes connected with cairns and sometimes not.
At Largie-beg, near Dippen, on a narrow plain near the sea,
several cairns lately opened contained rude urns of unbaked
clay, containing ashes." These cairns on the shore have, how-
ever, most probably a different history, having been raised
over the dead given up by the sea. "An urn containing human
bones," the writer continues, "turned up last year at the
manse [of Kilbride]; and at S. Kiscadale, near Whiting
Bay, there was found a few years ago an urn containing a
piece of gold in the form of a drawer-handle, with some iron
or steel much corroded at each end. The man concealed his
prize, and took the earliest opportunity of disposing of it to
a jeweller in Glasgow, who melted it down into rings and
brooches." — Pennant describes the Dun-fion pitchstone, but
takes no notice of the wall of the supposed fort on its
summit, or of any sign of vitrification ; there seems then no
foundation for this but the conjecture of Headrick already
alluded to (Art. 41). "In the course of my ride," says Pen-
nant, " on the other side of the hill of Dun-fion, facing the
bay of Lamlash, I saw on the road-side a cairn of a different
kind to what I had seen before; it was large, of an oblong
form, and composed like the others of round stones; but
along the top was a series of cells, some entire, but many
fallen in ; each was covered with a huge flat stone of great
size resting on others upright, that served as supports; but
I could not count them by reason of the lapse of the lesser
stones. . . . These cells are called in Wales Cromlih,
and Cest-va-en or chests." This cairn and cromlech seem no
longer to exist ; the stone circle at the summit level of the
road, and mentioned by us farther on, can hardly be indicated,
and no notice is taken in the Statistical Account of any such
monument in this neighbourhood. Doubtless it has shared
the fate which has overtaken so many of the ancient remains
both here, in Bute, and indeed over the whole country.

99. The most striking of the monuments which remain, and
the best preserved of the kind in Scotland, except those in

ANCIENT REMAINS IN ARRAN. 209

Orkney and the Isle of Lewis, are the Stone Circles of Tor-
more. They are situated on the open moorland at the back of
Drumadoon, and the King's Hill, S. of Mauchrie water-foot.
They may be most readily reached from Brodick, by crossing
the moor from the base of Craig-Dhu, S. of Glaister farm, or
by turning off the Shiskin road at the bridge over the
Mauchrie, passing along the Dougrie road through the old
birch wood as far as the first farm-house, where the weird
forms of " the standing stones" are well seen upon the moor.
If we ascend from the shore after visiting the King's Cove,
then a little beyond the farm-house of Tormore a large circle
of granite blocks is first passed ; a little farther, two upright
stones ai'e seen upon the moor, and when the eastern crest is
reached, the group of circles is seen on its eastern slope.
The tall, upright stones look majestic and solemn, standing
solitary on the wild heath. We associate them with an age
and the life of a people long passed away, but who were
moved by feelings and passions like our own, and they thus
awaken in our minds a strong human interest. The mystery
of their origin and purpose enhances the charm that hangs
about them ; curiosity and speculation are excited : and we
conceive a strong desire to discover for what purpose they
were reared, and to what uses they were applied. After a
careful survey of them, in July, 1860, and an examination of
all that had been written about them, I was not able to
satisfy myself that there was any authentic record of the
discovery of relics which could throw light on their origin or
use; while I was acquainted with only one or two cases in
which such circles had been anywhere opened. Further
investigation into a subject confessedly so obscure seeming
thus highly desirable, I placed the nature and importance of
the inquiry before the late Duke of Hamilton, and requested
permission to make a series of excavations within and around
the stone circles and monoliths of Arran. His Grace not
only at once most kindly acceded to the request, but ex-
pressed a wish that the operations should be conducted at

p

210 ANCIENT REMAINS IN ARRAN.

his expense. His Grace also placed me in communication
with his agent in Arran, Mr. Paterson of Whitehouse, Lam-
lash, whose judicious arrangements mainly contributed to
the success and rapid conduct of the operations. But
before giving an account of the excavations, it will be
necessary to describe the circles in their present state. Mr.
Paterson was so good as to have an actual survey made of
them, by which their relative positions are accurately laid
down by the compass, and the respective distances apart, to
a scale. These are shewn in the annexed plan, on which
also I have indicated the kind of circle, whether of pillared
stones or blocks.

The more eastern circle of the group (No. 1) is a single
circle of granite blocks, having a diameter of 14 yards; two of
the stones are entire, and stand about 5 feet high ; the others
merely protrude from the surface of the moor. The second
(No. 2) is a single circle of tall sandstone slabs, of which
three ("W. to N.E.) are perfect and upright; but the rest of
the circumference is defined by the bases of other stones
remaining in the soil — the number being, in all, seven or
eight. The pillars are from 12 to 14 feet in height, 3 to 4
feet broad, and 11 to 22 inches thick. Two circular slabs,
4^ feet in diameter, and 1 1 inches thick, plainly cut from a
fallen pillar in modern times, and intended for millstones, lie
inside the circle — one of them is pierced by a grooved
perforation, as if for the admission of a shaft. The next
circle (No. 3) is 13 yards in diameter; in the circumference
nine stones are distinctly seen, but one only is upright and
entire; its height is about 14 feet, and other dimensions
about the same as in the pillars of No. 2. Near the centre
there is a large square stone, crossed by five deep grooves.
The circle No. 4 is formed of four blocks of coarse granite,
standing nearly on the cardinal points, and about 3 feet high.
The figure is not a perfect circle, but rather an ellipsis whose
greater axis, directed north and south, is 7 yards long. The
circles now described are on that part of the moor which is

*

*4

4

* ^f-

•3

<f*\$

a gf^

^.^

ANCIENT REMAINS IN ARRAN. 211

covered with peat moss, and peat is cut from the banks
which are fast approaching the bases of the stones ; but this
will not affect their stability, as the peat is no more than 15
to 20 inches thick, and the pillared stones have a firm hold
in the red till bed below the peat : the peat, indeed, is most
probably the growth of a recent time.

The circle No. 5, a little to the west of No. 4, is on higher
ground, dry and gravelly, and has remarkable features, which
have led to its frequent mention, while others have been
overlooked. It is alluded to under the name of " Suidhi-
Choir-Fhion," or Fingal's Cauldron-Seat; the encircling
stones being, according to tradition, the support of the
giant's cauldron. ' It consists of a double circle of granite
blocks, boulders from the northern mountains, gathered from
the adjoining moor. The outer circle has 14, the inner 8
blocks, larger than those in the outer, and 3 to 4 feet high ;
they stand 3 to 4 yards apart in the inner row, the diameter
of which is 11 yards; the ring between the two circles is
from 5 to 7 feet broad, probably because the constructors
were in no way particular, or some of the blocks may have
been shifted. A block on the south-east side of the outer
circle has a ledge perforated by a round hole, which is well
worn on the edges; and, according to the usual tradition, is
said to have been formed for the purpose of fastening the
favourite dog Bran, belonging to the giant before named.
No. 6 is an enclosure by the road-side, a short distance W.
of No. 5, like an open raised grave; it is formed of five slabs
of sandstone placed on edge, 2 to 3 feet above the surface,
and 89 even all round that a large slab may once have fitted
on as a lid; the enclosure is 6 feet E. to W., by 4 feet
broad ; two slabs form the east side, one each of the others.
Nos. 7 and 9 are monoliths, but most probably there were
circles here also, as there are indications of other stones ; the
height is about 5 ;feet 6 inches. South of these is No. 8,
having several slabs, nearly the same size as the last, indicat-
ing a former circle. No. 10 is a large circle of granite

212 ANCIENT REMAINS IN ARRAN.

blocks, diameter 21 yards. It is close to Tormore farm-
house, and the most western of the group.

With regard now to the nature of the stones of which
these circles consist, it is worthy of note that — while the
granite circles consist of such loose blocks of the coarse and
fine varieties as are commingled on the moor, and could be
readily rolled into position by a number of men with crow-
bars, none of them being heavier than from 4 to 5 tons — the
pillared stones are of Old Red sandstone, which does not
occur nearer than the cliffs towards Auchincar, and are of
much greater weight, the largest being not less than from 8
to 10 tons, and the others 6 to 8 tons. The intervening
country is rough and difficult; yet there does not appear to
us any other conclusion possible than that this is the origin
whence they have come. All of them have plainly under-
gone a certain amount of coarse " dressing," but with what
tools it is impossible to say.

100. The excavations were begun on the morning of the
24th May, 1861, with a body of nine men, under command
of the chief hedger, whose intelligence and zeal were of the
greatest use to me. I broke ground in the circle No. 2 ;
and considering that the centre of the circle and the base
of a pillar were the spots most likely to receive any object
valued or venerated, I found the centre easily, knowing
three points on the circumference, and excavated there, and at
the base of the most southern of the three pillars, the one on
the S.W. side. Nothing was found here; but I may mention,
for the guidance of future excavators, that, except iu a peaty
or gravelly soil, the ground can be felt to a considerable depth
by means of a strong pole, armed with an iron facing, and
thus either much digging be saved or the right direction
given to it. The ground was sounded in this way, in the
central digging, after 15 inches of peat were cleared off and
the till reached, and we became aware, by the peculiar
sound, that we were over a flat stone of considerable size.
The interest of the inquiry now rapidly increased ; and when

ANCIENT REMAINS IN AERAN. 213

the earth was cleared off, and a large slab reached, the ex-
citement of our party was wound to a high pitch. There
could be no doubt we were upon a place of ancient sepulture,
and curiosity was highly excited to discover the contents.
The slab being 13 inches thick, was of great weight,
and no slight effort, even of our large party, was required to
raise it. As it slowly rose, a small, neat chamber was dis-
closed, but neither skeleton nor bone — nothing but an urn,
with a handful of black earth at the bottom; and these
" ashes of the urn " we regarded as all that remained of some
great chief, in whose honour and for whose last resting-place
these huge monuments had been reared around the central
cist. The urn was in excellent preservation, but of rude
construction. The cist was 3 feet long, 26 inches deep, and
22 inches broad, formed of four flat sandstone slabs, set on
edge, the bottom being of the natural hard till soil. In the
bottom, lying loosely about in black earth, four flint arrow-
heads of rude construction were found. The lid was 2 feet 2
inches from the surface, and fitted nicely on to the sides all
round. The length of the cist was nearly N. and S.

The circle No. 3 was next tried at the centre, and a similar
cist and urn found at a depth of 4 feet 6 inches from the sur-
face ; in the cist and soil over the lid several flint arrow-heads
were met with, of rude construction, and the urn was even more
rude than the other. The opening was now continued south-
wards from the centre, and another cist found at a higher
level, the surface of the lid being only 20 inches from the
surface, the cist itself 2 feet 9 inches; the slab was 4 feet
square, and 13 inches thick. No sooner was the northern
edge of this ponderous lid slightly raised, so as to give a
glimpse of the interior, than the interest of the party was
again raised to the highest pitch. A white object, like a
blanched human skull, loomed out from the deep obscurity of
the cist. We had come at last to a veritable human grave.
The skull proved to be tolerably perfect, with most of the
teeth entire, the upper jaw partly decayed, the lower only

214 ANCIENT REMAINS IN ARRAN.

traceable on the floor in outline ; at the opposite or north end
of the cist, a few long bones lay on the floor, and also two
flint arrow-heads. The cist was 3 feet long, 2 feet deep, and
16 inches broad. The dimensions, nearly the same in all
the cists, shew that these were not intended for depositories
of the body in an extended position. When cremation and
the urn were not employed, the body was interred in a con-
tracted position, sometimes on the side with the body and
limbs bent; sometimes in a sitting posture, the knees drawn
towards the chin, with the arms used in battle lying on
either side, ready to be grasped again when the dry bones
should be clothed with muscle, and the re-animated body
should start into new life!

The double granite circle No. 5 was next tried, but n>
thing was found save the two sides of a cist. The place had
obviously been disturbed before.

The small open enclosure No. 6 was also searched, but no-
thing met with except a fragment of a flint implement. An
unsuccessful search was also made at the monolith on the
moor — No. 9 in the plan — and this concluded the labours of
our first day; it was bright and warm, and our operations
were prolonged late into the evening.

101. My next examination was made on the 26th Septem-
ber, the force, and its organization by Mr. Paterson, being the
same as before. His Grace the Duke of Hamilton honoured
us with his presence during the greater part of the day,
remaining on the ground till nightfall, and aiding us by his
advice. The day was one of the most brilliant and genial of
the season ; the atmosphere had that unusual transparency,
ominous of change, which gives magical effects to a landscape.
The granite peaks were defined against the deep azure of the
north-eastern sky with wondrous distinctness — they seemed
close at hand, and lifted up into the clear air far beyond their
usual height.

As the circle No. 3 had been found to have a cist to one
aide of the centre, I was desirous of discovering now whether

ANCIENT REMAINS IN ARRAN. 215

the circle No. 2, already found, like No. 3, to have a central
cist, had also one removed from its centre. The result
shewed that it was so; a very perfect cist was found N. of
the centre, at 37 inches from the surface; but no object
whatever was in it, neither urn, bone, nor arrow-head, nor
did it appear to have been ever disturbed; indeed, it is cer-
tain it could not have been. The conclusion seems warranted
that it had been prepared as the others for a place of sepul-
ture, but never used. It was of about the same size as the
others, and like them lay nearly N. and S. To the south of
the centre there was no trace of a cist, and it seems there-
fore improbable that the practice was to form a series round
the central one.

Satisfied thus far with the result of the inquiry as regarded
the circles of upright stones, I was anxious to determine
whether a like arrangement existed in the case of the circles
of granite blocks. The cist at the centre appeared to indi-
cate the purpose for which the circumference was reared : and
if the central cist exist within the granite circles, the purpose
will seem to be the same for both. An imperfect cist had
been found in one — two granite circles remained to be
examined. The larger of these two, the most eastern of the
group, No. 1 on the plan, was tried at the centre, and along
a radius, but without success. Nothing was found, although
the usual depth was reached. In the other granite circle,
No. 4, we were more successful ; a beautifully perfect cist was
found, 40 inches below the centre, of the usual dimensions,
and lying as before, nearly N. and S., or almost exactly
N.N.E. and S.S.W. It .contained an urn in fragments,
some bone fragments under the urn, three rude arrow-heads,
and a small bronze pin, much decayed. This last object
added immensely to the interest of the inquiry; while
the central cist shews a like purpose in both structures —
granite blocks and pillared slabs, alike disposed in reference
to a central place of sepulture.

An excavation at the imperfect circle No. 8, high on the

216 ANCIENT REMAINS IN ARRAN.

moor, gave no result "whatever; the position of the centre
and course of the circumference were in this case, as in that
of No. 9, mere matters of conjecture. This excavation ex-
hausted our day; it was nearly dark when we left the
ground, and there was now no time left to examine the last
circle, No. 10. I regretted this the less, that it had a tossed
appearance as if it had been opened in more than one place,
perhaps in the hope of finding treasure concealed in a spot
round which there hung a certain amount of sanctity.

102. Mr Paterson having organized a force at Brodick
under the chief forester, next day was given to the supposed
site of a circle in Glen Shant, but without any result. The
small granite circle and pillared slab at the summit-level of
the Lamlash road (Art. 70) were then examined. Here, within
the circuit of the four blocks, and at a small depth, I found
a cist 26 inches long, 11 wide, and 10^ deep, cut out of the
solid sandstone rock, and fitted by a lid. In it there were
bone fragments and black earth, in the soil over it some rude
flint ai-row-heads. At the standing stone we found nothing.

The day turning out very wet and stormy in the after-
noon, I was unable to carry out my plan of examining some
of the monoliths about Brodick — as those on Mayish farm,
and that by the road-side at Invercloy. It would be
interesting to determine whether they ai-e true monoliths,
or the remains of circles; whether monumental, or com-
memorative of a battle or the judicial combat. Human
remains, or those of weapons, would mark these purposes;
the non-existenoe of any remains would probably merely
indicate a boundary or be commemorative of a treaty of
amity — that neither party " would pass this pillar for harm"
to the other.*

103. All the objects found during the excavations were
presented, by the desire of his Grace the late Duke of
Hamilton, to the Scottish Antiquarian Society, and may be

* Genesis xxxi. 51, 52.

ANCIENT REMAINS IN ARRAN. 21 /

seen in one group in their Museum at the Royal Society
House, on the Mound, Edinburgh.* The principal objects are
given on our two plates. The nature of the human remains
was felt to be of great importance in the inquiry — of what
race or type of head, of what sex and age, of what bodily
proportions, those of a warrior chief, or a tender female, was
the individual to whom they belonged ? The determination
of these questions must have an important bearing on the
purpose for which these huge works were erected. Anxious
to have the opinion of the highest authority upon these
questions, I submitted the entire of the remains for inspec-
tion to Dr. Allen Thomson, Professor of Anatomy and
Physiology in the University of Glasgow, who kindly
furnished a full report upon them, which is given at length
in the paper already referred to. The remains were also
examined by Professor Goodsir, Dr. Struthers, and Mr.
Turner, of Edinburgh. The skull is of the old British type,
and, so far, the evidence is in favour of a high antiquity for
these works ; it seems to be that of a young female or slender
male, just arrived at maturity; perhaps " the daughter of the
tribe," or the future chief, who it was hoped would one day
lead in the field of battle and in the chase. The teeth form a
nearly complete set; in most of them the enamel, though
brittle, was entire, but the bony part crumbled into powder
under a very slight pressure; this part was also of a dark
brown colour, as if partially charred; the wisdom teeth had
either just passed through the gum or were about to do so.
Some of the bones were human rib-bones, others shaft-
portions of human thigh bones; there were long bones,
portions of deer's horns; and also portions of an animal's
under jaw, most probably a dog or seal, but very difficult to
determine from the state of decay.

* By desire of his Grace likewise, a memoir on the subject, of
which the account here given is an abstract, was drawn up for the
Antiquarian Society, read May 1862, and published in their transac-
tions for 1863.

218 ANCIENT REMAINS IN ARRAN.

The black earth of the urns was most kindly examined
for me by Dr. Thomas Anderson, Professor of Chemistry,
Glasgow University. It contains many minute bone frag-
ments, small pebbles, sand, and ordinary soil, but no trace of
animal matter. The bones, carefully cleaned, have the
following composition : —

Phosphates, . 84 '11

Siliceous matter, . . . . ' . 6 '29

Organic matter, . . . . . • . 3 '57

Carbonate of lime, . . . . . .. 3 '41

Water, ,. 2'62

100-00

From this analysis there can be little doubt that the
bones have been burned, and that the earthy matter is
part of the soil introduced along with them when they were
gathered from the spot where the ceremony of incremation
was performed. The absence of nitrogen shews that the
organic matter is not of animal origin, but that it and the
siliceous matter are part of the soil which could not be
separated from the bones before analysis.

The urns are of the earliest forms, fashioned by the hand
before the potter's wheel was used in these western lauds;
the irregularity thence arising is very perceptible under the
lower band; the form of the mouth also is not perfectly
round. The ornamental markings are made by sloping
lines, neither straight nor parallel, and appear as if marked
in the soft clay with a bit of twig, the streak coming off
light towards the lower part ; the two rows of alternate dabs
appear as if made with the thick end of the twig; the lines
separating the bands of ornament are drawn without any
attention to accuracy. The height is 7| inches; diameter, 7
inches at the mouth, 3^ at the bottom ; circumference at the
lower baud, 25 inches; at the upper, 24 inches.

The urn was either sun-dried, like the Mexican adobes,
or subjected to but a very slight degree of burning, as Dr.

ANCIENT REMAINS IN AREAN. 219

Anderson finds that the external surface only is burnt, and
the interior appears to have been scarcely heated, for it still
contains upwards of 7 per cent, of the water of combination
of the clay, which would have been expelled had the urn
been strongly heated throughout. The clay has not been
brought into a uniform plastic mass, but has the structure
of a number of small pellets, with bits of stony fragments,
which Dr. Anderson suggests may have been used to prevent
the urn from cracking during the process of drying and
burning. He also considers that the reddish colour of
the surface, produced by a peroxidation of the iron, is an
additional proof that the urn was subjected to a slight
artificial heat.

The flint of which the implements consist is not found in
Arran, but it is met with on the opposite coast of Antrim,
where it forms beds in the chalk; and thence the rude stone
may have been derived, as Cantire and the islands had inter-
course with the north of Ireland from the very earliest
times. But this is part of a very wide question, and need
not be here dwelt upon.

The bronze pin is 2J inches long, with a polished surface
like that of a smooth bone, and resembles the pin of a brooch
or bracelet, but slightly thicker. It was chemically tested
by Dr. Stevenson Macadam, at that time Lecturer on
Chemistry, Edinburgh, and by this means, and a compari-
son with similar objects in the Museum of the Antiquarian
Society made by the Secretary, distinctly proved to be
of bronze, though much altered in structure.

104. From the facts which have now been stated, the
following inferences seem legitimately deducible : —

1. Whatever may have been the state of civilization
among the constructors of these works, a certain sense of
fitness or congruity must have existed in their minds ;
for though there are so many circles, thei*e is no mixture of
dissimilar stones — they are either all of sandstone or all of
granite.

22U ANCIENT REMAINS IN ARRAN.

2. They must have been capable of using mechanical appli-
ances of great power, eveu on the supposition that the
pillared stones are from the carboniferous formation, and
were fetched from the nearest point possible — the bed of
Mauchrie Water.

3. Archaeologists divide the pre-histoi'ic period in our
islands into the Stone and Bronze periods. If their classifi-
cation be correct, and if it be conceded that there is human
progress in every period, then the use of rude flint imple-
ments and of implements of bronze ought to be separated by
a wide intei-val, and only flint implements of the most perfect
forms, if any, ought to be associated with an article of
bronze. But in the present case, flint implements of the
rudest forms are associated with an article of bronze; the
two have co-existed, have been in use together, and thus the
Stone and Bronze periods have interlaced deeply; probably
more deeply in an isolated situation such as Arran, than on
the adjoining continent of Britain, where improvements in
processes of art would spread more rapidly. But we must
guard against attaching too much importance to a single
case. It is, however, highly desirable that inquiries of this
kind should be multiplied in order to test the truth of the
theory; it would be greatly invalidated if such association
were found to be frequent.

4. All the cists have their greatest length between N. and
N.N.E., and their construction may therefore be inferred to
have been anterior to the earlier Christian times in this
country, when a superstitious regard began to be cherished
for a direction pointing east The present amount of varia-
tion of the compass has been in all cases allowed for; and it
does certainly seem strange that the directions should so
agree towards a north point — they lie, in fact, roughly N.
and S., being all a little E. of N., the direction having
clearly no sort of connection with the inclination of the
surface of the ground, for though this is very various the
direction is always the same. Knowing the curious

ANCIENT REMAINS IN ARRAN. 221

astronomical relations made out by Sir John F. W. Herschel
in regard to the pyramids of Egypt, I took the liberty of
consulting him on the subject of the direction of the cists
within the Arran Stone Circles. He kindly replied that he
did not consider that it could have any connection with a
past conformation of the heavenly bodies, as affected by the
causes of change among them. To refer it to a past amount
of variation would of course be absurd. Shall we rather say
then that the direction was roughly taken N. and S. ;
that it had reference to the mid-day sun, or to a native
home of the race to which the constructors belonged, amid
the wilds of the north ?

5. The information furnished by the human and other
bones is not of a very definite kind; the human remains
belong to a young person, most probably a female. There
are portions of deer's horns, and of an animal's jaw, most
probably a dog or seal. The absence of the wanting portions
of the human skeleton is not easily accounted for. There
was no trace on the floor of the cist of such an amount of
matter as the decomposition of the other bones would have
left; but if once decomposed the matter may have been
removed, or absorbed by the soil through the floor or spaces
between the stones forming the sides, during successive
floodings of the cist with water from the soil above. It is
highly probable that such removal may have taken place;
while the form and position of the skull would preserve it
from decomposition for a much longer time. Had the
remains belonged to an aged or powerful male, we should
have had much stronger evidence than we now possess for
regarding the circumference of huge pillared stones as reared
in honour of a great chief or warrior, instead of being a place
of mere ordinary sepulture.

6. That these stone circles were erected for the object just
stated is further confirmed by the circumstance, that the
centre of each circle is marked by a cist; that the circumfer-
ence has been reared in all cases in reference to this central

222 ANCIENT REMAINS IN ARRAN.

cist, as well in the case of the pillared stones and granite
circles at Tormore, as of the singular little cist, with its four
encircling blocks, at the summit-level of the Lamlash road.
The limited total area of the circles is no objection to this
conclusion, as we know that the sites of many circles, which
have existed to a late time, are not now to be found. The
circles may have been applied later to other purposes, as of
worship, of judicial combat, or of meeting on great public
occasions, while intended in the first instance by their con-
structors as sepulchral monuments, marking off the sacred
precincts where lay the ashes or bones of the dead. The
reasoning, it is true, might be reversed, and it might be
argued that the circles were reared for religious or judicial
purposes, and afterwards adopted for sepulture, as venerated
places. But it would, of course, be necessary to produce the
evidence for such an original purpose, in order to do away
with the foi'ce of that here given. This, however, has yet to
be done. Antiquai-ians have too often recourse to hypotheses
which are far-fetched, and with which the wonderful, the
grand, and the terrible are associated, while simple natural
explanations are overlooked or rejected. A Druidical origin
has been assigned to these works; but there is no evidence
that the Druidical priests, or the rites of that worship, had
at any time a footing in Scotland. As regards the south of
England, Brittany, and other districts where remains con-
sidered to be truly Druidical exist, the soundness of the
mode of reasoning just indicated would be tested by a careful
examination inside the circles, on the plan described in the
foregoing account. To what extent such an examination
may have been made, I am unable to state; within Stoiie-
henge, one trial has, I believe, failed to discover any human
remains. Such are, howevei-, found abundantly in the
barrows and other earthworks on the adjoining plain.

223

FOSSILS OE THE AEEAN EOCKS.

105. THE following is a list of all the fossils yet disco vered
in the various sedimentary formations of Arran, corrected up
to the present time. Many of the best fossils in the list are
from the collections of Mr. John Young, Hunterian Museum,
Glasgow University, and Mr. James Thomson, an active
member of the Glasgow Geological Society. Mr. Nelson
Mitchell and Mr. Thomas Chapman have kindly supplied
several good species. The Corriegills' Orthoceras could not
be specifically determined, on account of the change produced
by a trap dike on the sandstone in which it was imbedded.
The list of Arctic shells was kindly supplied by the Rev.
Robt. Boog Watson of Edinburgh, their first discoverer, who
also, in the most prompt and kind way, placed at my dis-
posal all the information in his power in regard to the
localities whence the shells were obtained. The Rev. H. W.
Crosskey selected for me the Foraminifera found in the shell
bed, and forwarded them to Prof. Rupert Jones, to whose
kindness I am indebted for the names of the species.

THE SLATES.

No trace of any fossils.*

THE OLD RED SANDSTONE.

Markings on the South Sannox beds look like fucoid or

* Yet, if the views of Sir K Murchison in regard to the structure
of the central Highlands be correct, these slates must be regarded
as of Silurian age. Metamorphic action may have obliterated all
trace of fossils.

224

FOSSILS OF THE ARRAN ROCKS.

annelid impressions, but they do not seem to be decidedly
organic.

THE CARBONIFEROUS SERIES.

Planta.
Stigmaria ficoides (sandstone), .

Calamites cannseformis, „
Sphenopteris, (?)
Lepidodeudron obovatum, . .
Antholithes, sp.,

Sigillaria, sp., ....
Halonia tuberculosa, .
Lepidodendron, sp., .
Lepidophloios Vuenschiaaa,
Lepidostrobus (?) ambiguus,

— la t us, .
Vuenschianus,

Corrie, Salt-pans,

Laggan Bay.
Corrie, Salt-pans.
Salt-pans.

Laggan Bay.

Zoophyta.

Cyathophyllum Wrightii, .

expansuin,

Lophophyllum eruca, ....
Lithodendron junceum, . .
Stenopora tumidus, . » '; .

Echinodermata.

Archseocidaris Urii, ....
Poteriocrinus crasaus, . .
Hydreionocrinus globularis,
Crinoid stems, various,

Corrie, Salt-pans.

Corrie.

Corrie, Salt-pans.

Corrie, Salt-pans.

Salt-pans,
Salt-pans, Corrie.

Crustacea (Trilobites).
Griffithides mesotuberculatus, . . Salt-pans.

Polyzoa.

Ceriopora interporosa,
Fenestella, plebeia (?),

Salt-pans.

FOSSILS OF THE ARRAN ROCKS.

225

Fenestella, oculata, .

Morissii, .

Diastopoi'a megastoma,
Polypora papillata,

Salt-pans.

Brachiopoda.

Terebratula hastata, ....
Spirifera bisulcata,

lineata, ....

Urii, .

duplicicosta,

Spiriferina, octoplicata,

laminosa,

Athyris ambigua, ....

Royssii, ....

Retzia radialis, ....

Stropheniena rhomboidalis, var. analoga,
Rhynchonella pleurodon, .
Streptorhynchus crenistria,

Orthis resupinata, ....
Chonetes Hardrensis,
Productus giganteus,

Salt-pans.
Corrie, Salt-pans.

Salt-pans.

Corrie, Salt-pans.
Salt-pans.

semireticulatus,

— longispinus, var.

— latissimus,

— punctatus,

— costatus, .

— scabriculus,

— aculeatus,

— Youngianus (?),

Corrie, Salt-pans.
Corrie, Salt-pans, Glen Shirag,
Mauchrie Water.

Corrie, Salt-pans.

var. Martini,

lobatus, .

Salt-pans.

Lingula squamiformis,

Corrie.

Lamellibranchiata.

Myalina Verneuilii, .

Monomyaria. Dimyaria.

Corrie, Salt-pans.
Corrie, Salt-pans (?).

Aviculopecten fimbriatus, .
several undetermined.

226

FOSSILS OF THE ARBAN ROCKS.

Venus (?), two sp. undetermined,
Edmondia sulcata,
Nucula gibbosa,
Leda attenuata,
Conocardium aliforme,
Cypricardia rhombea,

Salt-pans.

Gasteropoda.

Pleurotomaria Yvanii,
Macrocheilus acutus,

Michotianus, .

fusiformis,

Naticopsis elliptica, .

variata,

Murchisonia striatula,
Trochus biserratus,
Loxonema scalaroidea (?), .
Bellerophon Urii,
striatus, .

Salt-pans.
Corrie.

Salt- pans.
Corrie.

Salt-pans.

Corrie.

Cephalopoda.

Orthoceras unguis,
attenuatum,

—T 0) • •

Nautilus ingens,
tuberculatus,

Corvie.
Salt-pans.

Corriegills, in sandstone.
Salt-pans.

Megalichthys Hibbertii,
Cochliodus magnus, .
Clenodus cristatus,
Cladodus, sp., .

Pisces.

Corrie.

Come, Salt-pans.

Laggan Bay.

Newer Pleiocene, or Arctic Sl&ll lleds.

Balanus crenatus. Modiola modiolus.

Panopsea Norvegica. Leda pygmoea.

pernula.

Tellina Baltica.
t'vprina Islandica.

Pecten Islandicus.

FOSSILS OF THE AEEAN ROCKS. 227

Astarte elliptica. Pecten opercularis.

Arctica. Littorina littorea.

compressa. Turritella communis.

— striata. Natica (?).

Cryptodon Sarsii.

Foraminifera and Entomostraca, in tlie Glacial Clay.

Rotalv Beccari.
Polystumella striato-punctata.

Cythere, a new species, closely allied to C. Lamarckiana, an
Atlantic living form undescribed.

228

THE FLOKA OF AREAN.

106. IN the preceding pages it has been shewn how fine a
field Arran offers to the geologist, as exhibiting many diver-
sified phenomena in a limited area. To the botanist it is
scarcely less interesting — such is the luxuriance and variety
of its vegetation, and such the rarity of some of the plants
contained in its flora. It is indeed true that scarcely any of
these are botanical treasures of the first order; still, there ai-e
several of unfrequent occurrence in the west of Scotland, and
many quite new and highly interesting to the naturalist
accustomed to the flora of England or the Lothians.

This richness of Arran as a botanical field is owing to two
causes, — its geographical position and the variety it affords
of situation and soil. Lying near the shores of the Scottish
Lowlands, and at the same time forming one of the Hebridean
chain of islands, it partakes of the flora of each region — the
common plants of its fields, woods, marshes, and road-sides
belonging chiefly to the former — the maritime species to the
latter. These advantages of position it no doubt does in some
measure share with the coasts of the Clyde estuaiy generally,
and especially with Bute. In Arran, however, they meet most
completely ; and accordingly we find that no district of equal
extent in the west of Scotland can rival it in the number of
species. It is to this circumstance — its situation at the
junction of two dissimilar botanical provinces — that the pecu-
liar richness and variety of the flora of the island are mainly
to be ascribed.

Secondly, The geological structure of Arran impresses a
marked character on its physical geography, and gives rise to
the greatest possible variety of station and soil. We have lofty
and precipitous mountains, wide-spread mooi-s, small alluvial

FLORA OF ARRAN. 229

plains, hot and sheltered glens, damp woods, and sandy sea-
shores. We have every kind and degree of exposure, from
the wind-swept top of Goatfell to the warm hollows of Glen
Cloy, and this within the compass of a few miles. We have
modern fir-plantations and natural birch woods; cultivated
fields and hedge-rows ; wide stretches of peat- bog ; rocky pro-
montories and caves; open strands and sand-hills. Of soils,
too, in the stricter sense of the word, there is a notable
variety. The general division of earths into sandy, loamy,
clayey, and so forth, is loose, and for scientific purposes inac-
curate ; distinctions far more natural, as well as valuable, are
furnished by the nature of the rocks, whose decomposed
materials form the soil. In Arrau the variety of the geo-
logical formations produces a corresponding diversity in the
composition of the earths; and though we are no doubt still
greatly in the dark as to the influence of soil on the habitat
of plants, there can yet be little doubt that the diversity of
soils furnished by the granites, traps, porphyries, slates, sand-
stones, shales, and limestones, must exercise a powerful
influence on the flora. Whatever may be the importance of
these conditions generally, they certainly exist here in a
remarkable degree.

To these advantages of geographical position and physical
character is to be added the scarcely less important one of
climate. Lying off the west coast of Scotland, aud thus almost
in the Atlantic, Arran enjoys a copious rain-fall, while its
insular position preserves it from those extremes of heat and
cold which are so injurious to vegetation. Hence it is that
the climate, while in many parts quite cold enough for the
ordinary plants of Britain, and while the mountains rise high
enough to be a fit habitation for several alpine species, is yet
sufficiently mild and equable for the growth of some usually
found in more southerly regions. These conditions exist also
in the extreme point of Cantire; and accordingly we find
Campbelton enjoying the mildest winter in Scotland, and
many tender species flourishing there in the open air.

230 FLORA OF ARRAN.

The effect of climate on the flora of Arraii is twofold, — it
increases the number of species, and it imparts a general
vigour and luxuriance to the vegetation, which makes it
interesting to the lover of the landscape picturesque, no less
than to the botanist. Every one must have observed how
much of the peculiar charm of Arran scenery is owing to this
circumstance. The bright green of the fields and pastures,
the woods that fringe the shore and cling to the slopes of the
lower hills, add grace and soft beauty to landscapes that
would otherwise be severe and gloomy in their grandeur;
and while they pleasingly relieve the monotonous gray of the
granite mountains, serve to heighten by contrast the effect
of the bare crags and jagged peaks that rise behind them.
In spring or early summer, when the grass sprouts fresh on
the hill-sides, and the varied foliage of the trees still pre-
serves the delicate green of youth; or in July, when the lower
ridges are purple with the rich heather-bloom, this variety of
colour is finest — and this is the best time to see Arran
scenery in its perfection. In September the grass is parched,
the heather begins to wither, and a brown or gray tint pre-
vails over the whole landscape, little relieved even by the
red and yellow hues of the autumnal trees. While, again,
if we turn our attention from the general outlines of the
scene, and look more closely at its details, we shall find no
less cause to admire the effects of an atmosphere moist and
warm. To it is owing not only the abundant growth of
natural wood which clothes the glens and sea-cliffs, but also
the size and splendour of many of our finest shrubs and wild
flowers, such as the laburnum, the hawthorn, the foxglove,
the honeysuckle, and perhaps most strikingly of all, the luxu-
riant profusion of the whole fern tribe, from the stately
Osmunda to the lowly Hymenophyllum.

107. On proceeding to analyze the flora of a given district
— that is, to examine the internal relations of its constituent
parts, noting what species, or genera, or families are abun-
dant, what rare, and what character is thus imparted to the

FLORA OF ARRAX. 231

vegetation — we may regard it under two aspects, as illustrating
the peculiar featui'es of the tract itself, or as indicating the
relations of that tract to the surrounding regions — what may
be called its general botanico-geographical position. Glancing
at the flora of Arran from this latter point of view, we shall
observe some interesting facts. Writers on the botanical
geography of Britain class the plants of our country according
to several " types of distribution," to denote their geographi-
cal range and affinities. Thus many species, occurring chiefly
in the east and south of England, are assigned to the Germanic
type; others, most of them denizens of the Scottish High-
lands, are referred to the Scandinavian; while others again,
characterizing Ireland and the westerly coasts of Britain, are
grouped under the Atlantic type. Some few there are, found
only in Cornwall and the west of Ireland, whose affinities
are with the Spanish Peninsula, and especially with the
mountains of the Asturias ; and for these there is proposed a
Lusitanian type. Somewhat similar to the position of Britain,
if we may be allowed to compare great things with small, is
the position of Arran. Most of its common species — the
plants of the field, the road-side, the marsh — it has in com-
mon with the western Lowlands of Scotland, which stand to
it in the same relation that the continent does to England.
These Lowland plants form the bulk of its flora. They
include almost all those of common occurrence, as well as
several of the rai-er sort — such as Ranunculus Lingua, Helian-
themum vulgare, Epipactis ensifolia, Samolus Valerandi, the
Botrychium and Ophioglossum, Asplenium ruta-muraria, etc.
Under this class are embraced nearly all the species that
frequent the cultivated land, the marshes and streams, the
woods and pastures ; together with several maritime ones —
as SUene maritima, OenantJie Lachenalii, Calystegia Sold-
anella.

A second "type" discernible in Arran is that which we
might call the Highland. To this group belong the alpine
plants of the highest granite mountains — Salix herbacea,

232 FLORA OF ARRAN.

Thalictrum aJpinum, A IcJiemilla alpina, Cryptogramma crispa,
and others; several also occurring in elevated situations in
various parts of the island — as Rhodiola rosea, Oxyria renifor-
mis, Hynienophyllum Wilsoni; besides a few found in moun-
tainous regions, though at no great altitude — such as Cory-
dalis daviculata and Polypodium Dryopteris.

We have, thirdly, a considerable class occupying, as it were,
in Arran, the place which the Atlantic type holds in Britain,
including the plants peculiar to the west coast, and especially
frequent in the Hebridean chain of islands. Such are many
of the maritime species — Mertensia maritima, Brassica Mon-
ensis, Sedum anglicum, Raphanus maritimus, as well as Pin-
guicula lusitanica, Gymnadenia conopsea, Drosera anglica,
and Listera cordata — plants found in various localities through
the interior. This class includes many of the most interest-
ing and characteristic plants of the island — not a few of
them such as will be entirely new to the English botanist.

But the most curious feature in the botanical geography of
Arran is the occurrence in its southern extremity of several
species scarcely elsewhere to be found in Scotland ; belonging,
in fact, to the flora of central England, and here apparently
quite projected, so to speak, from their ordinary range. Of
these the most remarkable are Lathyrus sylvestris, Verbascum
thapsus, Imda Helenium, Althcea ojficinalis, and Carlina vu2-
garis. They all occur within the circuit of a mile, on the
warm southern face of the cliffs and steep alluvial banks that
front the sea at the extreme south of the island, near Benan-
head. No one who examines the locality can think it pos-
sible that they should have escaped from cultivation; and it
is scarcely less improbable that they should have been planted
there by the hand of man. The Lathyrus, Verbascum, and
Carlina are still abundant; the Intda, however, seems to
have been extirpated, if indeed the report of its existence was
correct It is certainly not to be found now, yet it is difficult
to see how the mistake could have arisen, as there is no plant
in the neighbourhood which would be readily mistaken for it.

FLORA OF ARRAN. 233

The Althaea has been found within the last few years, but is
now either extinct or very scarce. It is to be hoped that
botanical collectors who may visit the spot will spare the
beautiful Lathyrus, and refrain from extirpating it in what
is probably its only Scotch locality.

108. Having remarked thus on the external relations of
Arran to the botanical geography of the surrounding lands,
we might pass on to speak of its internal aspects, and notice
the several floras of the mountain, the glen, and the shore,
the moor, the wood, and the marsh ; in other words, to dis-
tribute the plants according to their respective botanical
stations. By such an examination, however, no new facts or
principles or peculiar interest would be elicited. A little
experience will enable the student to frame such a classifica-
tion for himself.

Most of the rare plants of Arran are to be obtained on or
near to the coast, some decking the bright sands, as the
Brassica Monensis, the purple Jlertensia, or oyster plant, as
it is called from the flavour of its leaves, and the lovely
Calystegia Soldandla, with its creeping stems and flowers of
delicate pink; some dwelling in the salt marsh and wet
grounds that lie between the old sea-cliff and the present
tide-mark, such as the Oenanthe Lachenalii, Triglochin mari-
timum, and pcdustre, the pretty little Glaux, the blue Aster,
and several others. Here, too, though not properly maritime
plants, we often find the handsome Parnassia, and delicate
Anagallis tenella, Samolits Valerandi, Orchis latifolia, and in
the drier spots, Geranium pratense and Eryikrcea linearifolia.
The bare rocky crags and promontories, which here and there
diversify the generally accessible coast of the island, are gay
with the brilliant white, yellow, and pink flowers of Sedum
Anglicum and acre,Silene maritima and Spergularia marina;
the succulent Cotyledon fixing its roots in the rock clefts, and
the glossy green of the sea spleen wort, Asplenium marinum.
The curious sea cliff which lines the coast of the island in
almost every part, marking the level at which the sea stood

234 FLORA OF ARRAN.

in some former age, is in most places thickly covered with a
natural growth of oak, ash, birch, hazel, and other trees, and
is kept moist by the numerous streams that trickle down its
face, or precipitate themselves in cascades from its edge.
The shade and humidity thus produced render its vegetation
luxuriant and varied; and we find many interesting species
growing on or near this line of irregular cliff, some hanging
from its wooded sides, some springing rank in the wet caves
that pierce it, and some inhabiting the stony and marshy
ground at its base.

Here, among others, occur Veronica montana, Hypericum
Androsaemum, Geranium sanguineum, Sanicula europaea,
Eupatorium cannabinum, Lycopus europaeus, Listera owtta,
Habenaria cJdorantha. Here, too, most of the Arran ferns
may be found — Polypodium Dryopteris and PJiegopteris,
Cystopteris fragilis, Aspidium or Lastrea recurva (foenisecii),
Aspidium lobatum and aculeatum, Asplenium marinum,
Hymenophyllum Tunbridgense and Wilsoni, and the magni-
ficent Osmunda regalia. In enumerating these plants, we do
not mean that they are all to be met with in any one spot,
but that they all occur in some part or other of the sea cliff,
while many abound through the whole of its long extent.
Those species which have just been mentioned, with the
addition of one or two rarities, such as Epipoctis ensifolia
and Thalictrum flavum, form in the main the sylvestral flora
of the island, which it is therefore needless to speak of more
particularly. Similarly, there is little to distinguish the
vegetation of the lower glens from that which has been
described as characteristic of the sea cliff and the woods, not,
at least, till the point is reached where the larger trees grow
scarce, finally giving place to thickets of birch or hazel, or to
the open expanse of pasture and moor. Here the aspect of
the scene is changed, and plants quite different attract the
attention of the botanist. The greensward is gay with the
purple Gymnadenia conopsea, as beautiful as it is fragrant,
the blue Jasione montana, Pimpinella saxifraga, Gentiana

FLORA OF ARRAN. 235

campestris, Erytltraea centaurium, Habenaria viridis and
albida, Orchis maculata, and many other handsome plants.
Nestling among the heather we may find Circaea alpina,
Listera cordata, with its slender stem and minute yellowish
flowers, the taller Galium boreale, and the tender green of the
oak fern, Polypodium Dryopteris. Rubus saccatilis here and
there trails its long stems over the stony ground, while the
viscid leaves of the sundew, Drosera rotundifolia and anglica,
the tiny cream-coloured flowers of Pinguicula Litsitanica,
and straggling yellowish-green stems of the little Lycopodium
selaginoides, mix with the moss that grows thick round the
margin of the springs and rivulets. Supposing the glen,
whose botany we have been describing, to be in the southern
division of the island, we shall, on ascending still higher, find
ourselves, after a stiff climb, on a wide expanse of undulating
moorland, covered by a thick deposit of peat bog, interrupted
here and there by a rocky hill-top, or the deep cut channel of
some mountain burn. These moors, varying in height from
900 to 1400 feet, have little to interest the botanist. He
may travel over them for a whole day without meeting more
than two or three species among the coarse grass and heather,
mixed with rushes and cotton grass, which clothe the surface
of the peat moss. Generally it may be said that the flora of
the higher grounds in Arran is inferior to that of the low
country. The granite mountains in the northern part of the
island rise quite into the alpine region, and are covered with
snow for seven or eight months in the year. Yet, when
compared with the mountain tracts of the central Highlands,
they will be found to possess few alpine plants. This fact
seems to be in some measure owing to that predominance of
bare rock over grass and heather, which gives them, at some
distance, the appearance of an unbroken mass of gray granite.
Besides the stony character of the mountain slopes, the thin-
ness and ungeniality of the soil furnished by the decomposed
granite, the absence in the higher regions of springs and
streamlets, and perhaps the very steepness of the loftier

236 FLORA OF ARRAN.

summits, may all contribute to render the alpine flora of
Arran comparatively scanty and uninteresting. But as this
poverty is hardly more marked iu Arran than on the other
Clyde mountains and through the south-western Highlands
generally, it ought perhaps to be chiefly ascribed to the
distance of these regions from that centre of distribution
whence we suppose the alpine species of Scotland to have
spread themselves. The commonest plants on the high
mountains are Saxifraga stettaris and Alcheniilla alpina; the
former is scarce on Goatfell, occurring more abundantly on
the heights round the head of Glen Sanuox; the latter is
very frequent on all the higher peaks, and covers, with the
graceful drapery of its silky leaves, the ledges of many a
granite precipice. Salix herbacea — the dwarf willow, whose
woody stem scarcely rises from the ground — is found on most
of the principal summits; Oxyria reniformis and Rhodiola
rosea grow abundantly in the rock clefts ; Circaea alpina and
Saxifraga hypnoides occur occasionally near the summits of
the southei-n hills ; the pretty little Tha.lictru.rn alpinum
may be found in many places, as on Ben-Varen, on Goatfell,
at the head of Glen Cloy, and on the summit of the pass
leading from Glen Rosa into Glen Sannox; Cryptogramma
crispa, the parsley fern, has been noticed in several spots,
rooting deep among the loose blocks of stone that strew the
mountain-side, while among the other cryptogamous plants,
Pott/podium Dryopteris and Phegopteris, Cystopteris fragttis,
Hymenophyttum Wilsoni, and several of the alpine Lycopodia,
may be enumerated as denizens of the glens and mountains.

Regarding the water plants of the island, those which we
find in its marshes and streams, there is but little to be said.
They are few in number, and not in any way remarkable.
The list is nearly exhausted by the names of Hypericum
dodes, Ranunculus lingua, Drosera Anglica, Littordla locus-
iris, Alisma plantago and ranunculoides, and several species of
of Potamogeton and Carex. Of true lake plants there are very
few; and this fact suggests the remark, how deficient Arran is

FLORA OF ARRAN. 237

in this element of the picturesque. Lakes there are several,
but, with scarcely an exception, they are placed in the high
bleak moors, far above the limit to which trees ascend, and
generally away from the higher mountains, so that they add
little either of beauty or of sublimity to the scenery of the
island. Loch Tanna, by far the largest, lies in a high and
gloomy plateau, strewed with blocks of decomposing granite,
among which the stunted heather barely supports its exist-
ence—itself as black and uninteresting a sheet of water as
any pool in the fens of Lincolnshire; the aspect of the whole
scene is one of utter wildness and desolation, without
grandeur. The absence, however, of this element of the
beautiful in Arran scenery is scarcely remarked, since it is
more than compensated for by the character which the sea
imparts to every landscape.

109. Having thus briefly indicated the plants which the
botanist will meet with in each region of the island, it is
scarcely necessary to prescribe for him any special excursion
or walks. These are best left to his own taste and con-
venience. It may, however, be not amiss, in a few conclud-
ing words, to direct him to the districts where his rambles
will be attended with most pleasure and success. Probably
no part of the island will offer to him so many interesting
species as the vicinity of Brodick, especially if he direct his
walks to the Corriegills shore, proceed northwards to Corrie,
or explore the tangled thickets and dripping rocks at the
head of Glen Cloy. Around Loch Ranza, too, several
excellent plants may be obtained; while, even if the tourist
be not botanically inclined, he will find in the exceeding
beauty of the coast an ample reward for his walk along the
lovely shore between Glen Sannox and Newton Point. The
western coast offers many striking scenes, and everywhere
commands noble views of the broad sound of Kilbrannan,
with the hills and glens of Cantire beyond; but its botany
presents little that is new to one who has already examined
the eastern part of the island; while the interior of the

23b FLORA OF ARRAJT.

country is occupied by undulating granite mountains, seldom,
except at Glen Catacol, assuming forms of sublimity or
beauty, and clothed with no vegetation beyond the grass and
heather that grow among the slowly-decomposing blocks of
gray granite with which the ground is strewed for miles.
Such is the aspect of the country — bleak, wild, unvaried —
from Loch Ranza to Loch lorsa and Dougrie, where the
slate and old red sandstone formations succeed the harsher
granite, and subside with gentle declivities into the alluvial
plain through which the Mauchrie Water finds its way to the
sea. Despite what has been said of the botanical attractions
of Brodick, there is no district in Arran that will better
repay the trouble of a visit than the south coast, from Slao-
dridh to Whiting Bay. Without any of the alpine grandeur
of the north, it has many striking beauties of its own — smiling
little bays, steep green banks, and bold cliffs of basaltic rock,
porphyry, or claystone, jutting far out among the waves,
or running in tall colonnades along the shore; seaward there
is the wide expanse of glorious blue, with the magnificent
pinnacle of Ailsa full in front ; beyond all, closing the distant
horizon, the gleaming cliffs of Ayrshire and the far-off coast
of Ireland. It is a delightful shore to wander along slowly,
searching and prying for rarities in the salt-marsh by the
water's brink, or up some leafy gorge through which the
streamlet from the hills forces its seaward way, forgetting
the world without and all its cares, delivered from the
dominion of dusty roads, and the necessity of getting home
in time for dinner. Let the naturalist take up his quarters
in the neat little inn that lies nestling in its snug little
hollow at Lag; there he may pass happy days in exploring
that solitary shore, and at night pull out into the deep, and
taste the unwonted pleasures of sea-fishing.

The mountains at the head of Glen Sannox will be found
richer in alpine plants than Goatfell, though scarcely equalling
it in height. Yet even their flora must appear scanty and
uninteresting to one who has botanized over the ranges of the

LIST OF PLANTS.

central and eastern Highlands, Ben-Lawers, Braemar, or
Clova. But at this he will have no cause to repine when, he
finds himself led into some of the most magnificent mountain
scenery in Britain. It has bean said that no scenery in
Scotland but that of Coruisk in Skye, and that of Glencoe,
can rival the grandeur of Glen Sannox; certainly neither the
Perthshire Highlands nor the English lake country can
shew anything so wildly sublime as the precipices and
aiguilles of rifted granite which tower around the heads of
those Arran glens.

LIST OF PLANTS.

110. The following list comprises only the rarer plants of
Arran, including under this title both those which, though
frequent in the west of Scotland, are seldom found in other
parts of the country, and those which, while abundant in
other parts of Britain, may appear somewhat uncommon to
one who knows only the botany of the west coast. It often
becomes difficult to draw the line, and to say what plants
should be omitted and what inserted. We cannot hope to
have decided always judiciously; when we have erred, it has
no doubt been in admitting plants which the practised
botanist will think too common to be placed in any but an
exhaustive list.

All the species of the fern tribe known to exist in the
island have been inserted, and their localities given in full
detail; partly because they bear a larger proportion than
usual to the flora of Arraii, and partly for the sake of those
who may take a special interest in them, or wish to procure
specimens for cultivation.

The genera are arranged according to the natural orders,
and the names those which appear in Hooker and Arnott's
British Flora.

Our best thanks are due to Professor J. Hutton Balfour of

240

FLORA OF ARRAN.

Edinburgh, for his kindness in placing his lists at our disposal.
On the authority of these many species have been inserted.
A few not observed either by Dr. Balfour or ourselves are
inserted on the authority of the late Dr. Landsborough. To
these the letter L is affixed. Plants which we suspect to be
not truly wild are indicated by an asterisk. It may be
added, that as such a catalogue cannot profess to be complete,
any hints for its correction or extension will be gratefully
received.

Thalictrum alpinum,

flavum,
minus,

( Goatfell,head of GlenSannox
Ben-Bharrain, head of
( Glen Cloy.

Whiting Bay(L).

Whiting Bay.

In many places.

Lamlash.

Ranunculus hederaceus, .

Lingua,

acris (vai\ pumilus) Near Lamlash.

sceleratus,

Trollius Europseus,
Corydalis claviculata,
Cardamine hirsute, .
Cochlearia officinalis

varieties),
Cakile maritima,
Lepidium Smithii, .

campestre,

Sisymbrium Sophia,

Brassica Monensis, .

oleracea,

campestris,

Crambe mai-itima,
Kaphanus maritimus,
Helianthemum vulgare, .
Viola tricolor (var. arvensis),

• palustris, .

Drosera Anglica,

rotundifolia,

Parnassia palustris, .
Silene maritima,

House roofs and woody places.
. Near Lamlash.
(several

On the shores.
. . Frequent on sandy shores.

Brodick.

Loch Ranza (?).
. Sandy sea shores.

f Sands at Brodick, Sannox,
\ Blackwater-foot, <fec.

Lag.

Mauchrie.

Imochar.

Southend.

Kildonan.

Marshy places.
Frequent.

Do.

Do.
Shores — frequent.

LIST OF PLANTS.

24V

Lychnis diurna, . Y

vespertina, .

Agrostemma Githago,
Sagina maritima,

nodosa,

Honckenya peploides,
Stellaria media,
Spergularia marina, .

Radiola millegrana, .

Althsea officinalis,
Malva sylvestris,
Hypericum perforatum, .

— dubium,

Androssemum,

quadrangulum,

humifusum,

— pulchrum,

hirsutum,

elodes, . • ^

Geranium pratense,

— sanguineum,

dissectum,

pusillum, .

Trifolium filifoi-me, .

medium, .

procumbens,

Anthyllis vulneraria,
Lotus major,
Orobus tuberosus,
Vicia sylvatica,

hirsuta, .

Cracca, .

sepium, .

Lathyrus sylvestris,
Geum intermedium,

urbanum,

rivale, .

Comarum palustre, .
Alchemilla vulgaris,
alpina, .

Common.
Do.

Cornfields — common.

On the shores.

Lamlash.

Sandy shores.

Frequent.

On the coast.

( Goatfell, Loch Ranza, Spring-
( bank.

Struey rocks (?).

Frequent.

( Invercloy, Whiting Bay, &c.,
I abundant.

Frequent in thickets.

Wet places.

Frequent.

Common.

Lamlash.

King's Cove and Loch Ranza.
( Brodick, Benan-head, Holy
( Isle, &c.

Struey, Thundergay, Dippen.

Frequent.
Do.

Frequent.

Common.

In the woods.

King's Cove, Kildonan, <fec.

Frequent.
Do. (?)
Struey rocks.
Frequent.
Common.
Frequent.
Frequent.
Common.
On the mountains.

242

FLORA OF ARRAN.

Alchemilla alpina (var. conjuncta), Glen Sannox.

arvensis,

Agrimonia Eupatoria,
Prunus communis, .
Rosa spinosissima, .

involuta, .

tomentosa,

villosa,

Pyrus aucuparia,

pinnatifida,

Rubus carpinifolius,

corylifblius, .

Idseus, .

saxatilis,

suberectus, . .,

glandulosus, .

rhamnifolius,

Circsea Lutetiana, .
Circsea alpina, . . .
Peplis Portula,

Ly thrum salicaria, . ,
Montia fontana,
Scleranthus annuus, . ,
Cotyledon umbilicus, ,
Sedum Rhodiola, . .
* Telephium,

acre,

Anglicum, . • ,

Saxifraga stellaris, .

hypnoides,

Hydrocotyle vulgaris,
^Ethusa Cynapium, .
Sanicula Europsea, .
Eryngium maritimum,
Bunium flexuosum, .
Helosciadium nodiflorum,
. do. (var. repens), .

Carum * Carui,
Pimpinella Saxifraga,
Oenanthe Lachenalii,

Frequent.

On the south coast.

Frequent.
Common.
Lamlash.
Dippen.

Frequent.

( Glen Eas-an-Bhiorach, Cais-
\ tael Abhael.

Lamlash.

Do.

f "Woods and mountains — fre-
( quent.

Head of Glen Cloy.

Holy Island and Lamlash.

Frequent.

Glen Cloy, Ben-leister Glen.

Frequent.

In many places.
Frequent.
Mountains — frequent.

Frequent on the shores.

Common.

Mountains — frequent.

Glen Cloy, <fec.

Common.

Waste ground.

Thickets.

Sandy shores.

Frequent.

Shore at Leac-a-Bhreac.

Lamlash.

Frequent.
Corriegills, &c.

LIST OF PLANTS.

243

Conium maculatum, .
Smyrnium olusatrura,
Scandix Pecten,
Myrrhis odorata,
Ligusticum Scoticum,
Apium graveolens, ..

Galium boreale,

Asperula odorata,
Valeriana officinalis,
Fedia dentata,
Cnicus pratensis,
Carlina vulgaris,
Carduus palustris,
Bidens cernua,

tx-ipartita,

Inula Helenium,
Eupatorium cannabinum, .
Antennaria dioica, >: .:
Cichorium Intybus, .
Gnaphalium uliginosura, .

sylvaticum, .

Aster Tripolium, .
Hieracium murorum, . ,

sylvaticum (var.

vulgatum), .
Filago German ica, .

• minima, " . . «.

Solidago virgaurea, .
Senecio sylvaticus, . .
Pulicaria dysenterica,
Matricaria maritima, .
Jasione montana, % »

Lobelia Dortmanna,

Vaccinium Myrtillus,

Vitis-Idaea,

Arctostaphylos Uva-ursi, .
Pyrola minor, .
Ilex Aquifolium,
Fraxinus excelsior, .
Erythrsea Centaurium,
linearifolia,

Dippen, Lag.

Dippen (L).

Frequent.

Dippen.

South End, Kildouan, <fec.

Shore at Lag, and Loch Ranza.
J Glen. Loig, North Sannox,
( and other places.

Woods.

Freqiient.

Loch Ranza.

Struey rocks.

Brodick.
Lamlash.
Struey rocks (?).
In many places.
Pastures and moors.
Whiting bay.
Loch Ranza.

Do.

Salt marshes.
Near Lamlash.

Glen Eas-an-Bhiorach.

Brodick, Glen Eais-dale.

Glen Easdale.

Glens — frequent.

Lamlash.

Struey rocks.

Sea coast.

Common.

J Small loch near Loch Ranza ;
( Loch lorsa.

Frequent.

On the higher mountains.
Holy Isle, Glen Eas-an-Bhiorach.

Holy Island (L).

On the cliffs.

Rocks at Glen Catacol.

Frequent.

Shore at Corriegills, &c.

244

FLORA OF ARRAN.

Gentiana campestris,
Menyanthes trifoliata,
Convolvulus arvensis,
Calystegia sepium, .

Soldanella,

Cuscuta *epilinum, .
Lithospermum officinale,
arvense,

Mertensia maritima,
Solanum Dulcamara,
Veronica montana, .

Melampyrum pratense,
Scorphularia aquatica,
Digitalis purpurea, .
Linaria vulgaris,
Verbascum Thapsus,
Lycopus Europaeus,

Moors and pastures.

Springbank, <fec.

Corrie.

Hedges — frequent.

Blackwater-foot.

Lamlash.

Loch Ranza.

( On the sands, Brodick, San-
\ nox, &c.

Brodick, Holy Island, <fcc.
( Blue Rock, and woods in
( other places.

Abundant.

Near Sannox.

Frequent.

Cliffs at Struey and Dippen.

Sannox.
Menthasylvestris (var. velutina),

• sativa (var. rubra), Lamlash and Whiting Bay.

viridis,

* rotundifolia,

Galeopsis versicolor,
Calamintha Cliuopodium (?),
Lamiura amplexicanle,
incisum (?),

intermedium,

Stachys ambigua, . . .

Scutellaria galericulata, .

Origanum vulgare (?), i

Pinguicula vulgaris, .

Pinguicula Lusitanica,
Utricularia vulgaris,

Near Corrie.

Brodick.

Frequent.

Glaux maritima,
Anagallis arvensis,

tenella,

Samolus Valerandi,
Armeria maritima,

Loch Ranza.

Do.

( Lamlash, Kildonan, Mauch-
( rie, «fec.

Slaodridh.

Brodick, Struey, «fec.

Abundant

( In the bogs, but now less fre-
( quent than it was formerly .

Loch-an-Davie, Loch lorsa.
J Loch on the shore, near
\ Mauchrie.

Salt marshes.

In cultivated ground.

Abundant in the bogs.

Marshes — frequent.

Sea shore — abundant.

LIST OF PLANTS.

245

Littorella lacustris, .
Plantago maritima, .

coronopus,

Atriplex laciniata, .

angustifolia,

do. (var.erecta),
Salicornia herbacea, .

Suseda maritima,

Salsola Kali, .
Rumex congloraeratus,

pratensis,

Polygonum Raii,

Convolvulus,

Oxyria reniformis, .
Empetrum nigrum, .
Parietaria officinalis,
Ulmus montana,
Calliti'iche pedunculata,
Myrica Gale, .
Betula alba,
Salix herbacea,
Corylus Avellana,
Populas tremula,
Juniperus communis,
nana,

Epipactis ensifolia, .
Listera ovata, .
cordata, ,

Orchis mascula, *•

maculata,

latifolia,

Habenaria bifolia, .
Chlorantha,

viridis, .

albida, .

Liparis Loeselii (?), .
Gyranadenia conopsea,
Malaxis paludosa,

Loch Ranza.

On the shores.

Dry places.

Sea shores.

Shore at Struey.

Slaodridh.

Loch Ranza.

On the shores — Loch Raiiza,

&c.
Sandy shores — frequent.

Lamlash.

Shore at Lamlash, Lag, &c.

On the shores.

Mountains.

Lag.

Brodick Castle.

Abundant.

Frequent.

Mountain tops.

Abundant.

Torlin.

Mountains.

Goatfell. (?)

Slaodridh, Invercloy, Whiting

Bay.
Lamlash.
Cior - Vor, Suidhe - Fergus,

Glen Cloy.
Abundant.
Do.

Do.

Do.

Do.
Whitefarland, Eas-a-mhor,

Loch Ranza.
Loch Ranza ; and dry heaths,

in many places.

Abundant in fields and heaths.
Kildonan.

246

FLORA OF ARRAK.

Allium ursinum,

*" Schoenoprasum,

Agraphis nutans,
Juncus triglumis,

trifidus,

compressus (var.

Gerardi),

glaucus,

maritimus, .

Narthecium ossifragum, .
Alisma Plantago,

- ranunculoides,
Triglochin pahistre, .

maritimum, • .

Typha lotifolia,
Sparganium ramosum,

simplex,

Potamogeton plantagineus,

oblongus,

Zostera marina,
Eriophorum vaginatum, .

polystachion,

augustifolium,

Schoenus nigricans, .
Eleocharis .palustris,

do. (var.

uniglumis),

multicaulis,

Blysmus rufus, .

compressus, .

Scirpus maritimus, .

pauciflorus, .

Carex ampullacea, .

Carex Isevigata, . .

vulpina,

pauciflora,

arenana,
remota,
curta,
divisa, .

Abundant.
Glen Shirag.
Abundant.
Goatfell.
Do.

I Salt marshes.

Torlin.

Sea shores.

Abundant.

Brodick, <fec.

In several places.

Frequent.

Salt marshes.

Mill-dam, Whiting Bay, Jirc.

Lamlash.

Brodick.
Loch-an-Davie.
Abundant.
Goatfell.

Abundant.
Sea shores.

Do.

( Sands near Kildonan, and at
I Lag.
On the coast.
On the sea shore.
Do.
Do.

Corregills and Corrie shore.
Near Mauchrie.
( Roadside between Brodick and
Lamlash : Loch Ranza,
( Corriegills.

Shores.

( Ascent of Goatfell, and in
( Glen Rosa.
Sandy shores.

On the moors.
Frequent

LIST OF PLANTS.

247

Carex distans, .
extensa,

fulva, .

— glauca, .

— muricata,

ovalis, .

Oederi, .

paniculata,

rigida, .

Rhyncospora alba, .
Isolepis Savii, .

'™ ' SO LtlCL'tl^ • •

Avena planiculmis, . . -j

Ammophlla arundinacea, .
Elymus arenarius, .
Briza media (?),
Aira alpina,

Brachypodium sylvaticum,
Catabrosa aquatica, .
Festuca broinoides, .

gigantea,

vivipara,

Molinia coerulea (var. alpina),
Poa maritima, .
Triticum junceum, .

laxum,

Polypodiurn vulgare,

Imochair.

Dippen.
Near Dougrie.

Goatfell.

North Glen Sannox, &c.

Corrie.

Frequent.

Said to have been found in

Glen Sanuox.
On the shores.
Sandy shores.

Goatfell. (?)

Woods.

Shore at Kildonan and Lag.

Lamlash.

Corriegills.

Goatfell.

On the mountains.

On the shores.

Shore at Slaodridh.

Sands at Struey.

Very common.

( Abundant in the woods, and
along the sea cliff; fre-
quent also in sheltered
Phegopteris, •{ spots in the glens; and
sometimes found in rock
crevices at the tops of the
highest mountains.

( In the woods, and very abun-
dant in damp sheltered
spots in the glens and

Dryopteris, among the heather; often

ascending the highest
mountains along with the
last species.

248

FLORA OF ARRAN.

Aspidium lobatum, .
aculeatum,

Oreopteris,

Aspidium spinulosum,

— Filix mas,

recurvum (Foeni-
secii), .

Cystopteris fragilis, .

Asplenium Ruta-muraria,
Trichomanes, .

• Frequent on cliffs and banks
near the sea ; sometimes
in exposed situations, pass-
ing into the form lonchi-
tioides.

Frequent in hedges and woods.

'Very abundant in the glens
and on the moors, often, as
in the lower part of Glen
Rosa, covering the whole
hill-side with the delicate
yellowish - green of its
young fronds.

' This variable and perplexing
species is very abundant
in several varieties, one of
which, with an ovate or
ovate-lanceolate frond, is
frequent on the highest
mountains.

Abundant everywhere.

(^Frequent on the sea cliff — as
at Corrie, North Sannox,
Salt - pans, and Whiting
Bay ; in rocky and wooded
ground at the head of Glen
Cloy, under the Sheeans.

( Rocks at the head of Glen
Cloy ; cave in the old sea
cliff at Blue Rock ; near
Glen Sannox; and in

J several other spots among
the mountains, as, near
the top of Ben-Ghnuis, on
the Ceims, &c. It gener-
ally, if not always, occurs
in the form dentata.
Old walls at Brodick Castle,
and probably in other
places.
Common on rocks and walls.

LIST OF PLANTS.

249

Asplenium viride,

marmum,

f Abundant on limestone cliffs
j at the head of Ben-leister
•j, Glen, above Lamlash ; on
rocks near the head of
Glen Cloy, sparingly.
('Occurring here and there in
caves along the old sea
cliff. It is now scarcely
to be found on the east
and north coasts of the
island, but may be obtained
in many places on the
south and west, as at
Benan-head, King's Cove,
and Iinochar. At Leac-a-
Bhreac, near Blackwater-
foot, we have gathered
fronds eighteen inches
long.
Adinatum nigrum, Abundanton banks and rocks.

TJ, . , . f • ( Very abundant and beautiful,

i ilix-Kemma J -, ,. . ~ . '

and presenting an infinite

( variety of forms.

J Hedges, banks, and rocks ;

( often very luxuriant.
Common.

Among granite blocks at the
head of Glen Sannox, under
the crags of Cior-Mhor,
abundant ; also on the pass
leading from Glen Rosa
into lorsa, in small quan-
tity, and probably in seve-
ral other places. One very
small plant we found on
one occasion at the sum-
mit of Ben-Ghnuis.
Common.

C Fields near the String road
in Glen Shirag; also on
the shore at Invercloy and
Corriegills, and in dry
pastures in other places.

(Athyrium Filix-foemina),

Scolopendrium vulgare,
Pteris aquilina,

Cryptogramma crispa,

Blechnum boreale,

Botrychium Lmnaria,

250

FLORA OF ARRAN.

Trichomanes radicans,

Hymenophyllum Tunbridg-
ense,

Wflsoni,

Has recently been discovered
in moist cliffs near Corrie.
It is in very small quan-
tity, and will be extirpated
if the visitors who are for-
tunate enough to find it
do not exercise consider-
able self-control.

On sheltered rocks in several
places, as in a plantation
near Invercloy, in the
wood at Brodick Castle;
near Corrie; at Fallen
Rocks; and at the entrance
of GlenEais-dale, Whiting
Bay. It is remarkable
that, in all these localities,
it is found growing upon
the old sea-cliff; as is also
the case where it occurs
atDunoon, Holy Loch, the
Kyles of Bute, and the
shores of Loch-Lomond
at Tarbet.

Abundant in many spots, as
near Corrie; sea coast at
Sannox; Glen Cloy; Ben-
leister Glen; Birk Glen,
above Invercloy, often
growing with the last-men-
tioned species, but also as-
cending to the tops of the
highest mountains, where,
as on Ben-Ghnuis, it covers
the most exposed rocks at
a height of 2500 feet, and
is with difficulty distin-
guished from a moss. In
general it prefers damper
spots than H. Tur^
bridgense, and stands ex-
posure much better.

LIST OF PLANTS.

251

Osmunda regalis,

Ophioglossum vulgatum, .

Lycopodium clauatum,
Selago,

annotinum, .

selaginoides, .

• alpinum,

Equisetum fluviatile (Telma-

teia), .
arvense,

sylvaticum, .

limosum,

Frequent, on the sea cliff, all
round the island, often
very luxuriant, attaining
the height of 10 or even
12 feet. It is now less
abundant than formerly
on the east coast, but may
be found in profusion near
Loch Eanza, and at King's
Cove.

!ln a meadow between Benan-
head and Torlin, and pro-
bably in other places.
Abundant on the hills.
(Moors and mountains — fre-
( quent.

Said, but in all probability
erroneously, to have been
found on Goatfell.
Boggyand springy places on the
} hills and glens — frequent.
On the mountains.

> Frequent in damp ground.

— palustre,

Common.

In the woods and glens, not

unfrequent.
Ditches and ponds.
Boggy places — frequent.

353

MARINE ZOOLOGY OF ARRAN.

111. THE coast-line of Arran abounds with objects of
interest. Xor is this interest confined to geological science.
The naturalist — and especially the student of marine zoology
— will find here a field for investigation ; and even the
casual visitor, if endowed with the faculty of inqnisitiveness,
may obtain instinctive and amusing occupation.

Let it be understood that the waters of the sea generally
teem with animated beings of an endless variety in form,
size, and structure, and that myriads of these creatures are
tikft mtofliilmlhi of rock-pools, where they share the vicissi-
tudes common to mortality — sporting, at one moment, in the
full enjoyment of the pleasures of life, and at another instant
either wrangling or wrestling with an antagonist — perhaps
for a mouthful of food— or " scudding " with all their might
from the open Jaws of an approaching and ravenous enemy.

We will TTi|HPf"" that we are now standing at Corriegills,
on. the sea nVorc in Brodick Bay, about one mile from In ver-
doy. Hie rocks am of the red sandstone formation ; and
the aoft material, yielding to the action of ceaseless tides, and
to die ladling waves of the winter storm, has been hewn in
all directions iajte fissures and small basins, which, renewed
with fresh supplies of water by each returning flood, are the
habitats of plants and MJIM!^ Look into these adjacent
pools. They appear to be almost covered with tiny sea-
weeds. Yet toe jungles of India are not more fully occupied
with their appropriate «faMi«nMM titan are these patches of the
SOL They abound with microscopic life, generated under
the »aisath of tike ami ; and, by stooping down and watching

MAEIXE ZOOLOGY OF AREAS. - "

for an instant, you will observe as modi bustle, enterprise,
and activity, with the usual accompaniments of snceem and
misfortune, as are characteristic of an opulent city. The
water at this moment is perfectly stilL There is not a,
breath of air to disturb its surface. Bat do yon notice that
ripple 1 Depend upon it there is some mischief not far off
See ! there is a prawn, or something like a prawn, swimming
with the greatest eagerness towards a piece of rock almost
concealed by marine vegetation. Nay, there is a whole fleet
of these creatures, trying to outvie each other la speed, and
all shaping their course in the same direction. With 40
help of our gauze or muslin landing-net we will endeavour to
catch a specimen. Skill and patience are required, for they
are uncommonly agile — they leap with the suddenness and
rapidity of a grasshopper, and they are too "knowing" to
enter a bag, however fine may be its fabric, unless they are
enticed or surprised by stratagem. Look how carefully that
individual is keeping his fact towards your not — see how
deliberately he "backs astern," with an occasional jump,
when you think that he cannot possibly escape : — but now,
keep your hand steady ; lie win approach - he will inspect
your net with aU imaginable «"q»*«BB*y Observe how skil-
fully he uses his "horns," or antennae— exquisitely finjamd
instruments, half as long again as his entire body, and which,
as with the insect tribes, are delicate organs of inquiry. He
wffl creep round the net and touch every part within his
reach. But stop — here is a chance of catching him— -run him
gently back into that nook in his JmmMK»fa» rear, dip your
net under him, and then, if you are smart in raising your
hand, he will be your prisoner. Bring him on shore, and,
inverting the net into a glass vessel of sea water, we wffl be
able to examine him at leisure. But what has become of
his companions I Whither were they bound! And on
what errand were they so swift 1 It is a case of piracy.
These crustaceans, which we saw swimming in such baste,
are a fleet of sea-robbers ; they are at this moment plunder-

2o4 MARINE ZOOLOGY OF ARRAN.

ing a colony of helpless zoophytes, and, as often happens
under analogous circumstances, they are quarrelling among
themselves. Look, first, at these two warlike specimens of
their race engaged in combat — each endeavouring to get at
the prey. They do not appear to use their eyes, although
these organs are well developed ; but, feeling about on all
sides with their antennae, they no sooner touch each other
than the strongest of the two rushes forward, assaults, and
beats back the weaker, which, after being foiled in repeated
attempts to hold his position and obtain a share of the spoil,
retires to another portion of the invaded colony. Now
observe their thieving propensities. Here is a group of sea
anemones, just below the surface of the pool, curiously fixed
with their base in a chink of the sandstone rock, whilst their
upper or anterior parts, fringed with tentacula, are exposed,
like the sunflower when its petals are expanded, to the
cheering rays of light. These are at present undisturbed.
Here is another group; and of these several have made prize
of some food, brought within their reach by the graceful
vibrations of their tentacula. The crustaceans — whether
they scented the dainty morsels or not, we cannot determine
— have hurried to the spot in order to appropriate as plunder
what has been legitimately obtained by others. Each prawn
is helping himself, drawing the coveted bits from the grasp
of the helpless and disappointed zoophytes : and see how he
scampers off with both hands full to a place of retreat But
look once more. You observe an anemone in the act of
swallowing a piece of decayed shell-fish — his mouth opens
wider and wider — and now the morsel has disappeared.
Here is a prawn just arrived — almost too late — he feels with
his finger and thumb about the lips of the closed mouth —
there is apparently nothing for him ; yet he seems to know
what he is about — and true enough ; for see how artistic-
ally, how professionally he introduces his hand into the
mouth of the anemone ! He will not be defeated. Nor has
he any compunction. He thrusts his hand down, and down,

MARINE ZOOLOGY OF ARRAN. 255

and yet deeper still, into the very stomach of the animal. He
has actually got possession of every particle that had been
swallowed ; and now he is off with all possible speed lest his
ill-gotten booty should be snatched from him by a prawn
more valiant than himself.

This colony of plundered zoophytes is formed by a species
of anemone common only on certain parts of the coast — it is
the Actinia bellis, described by naturalists. It sometimes
escapes detection, from the fact that almost the whole body
is often concealed within the chink of a rock, whilst the
expended disc and the surrounding tentacula are partially
obscured by the shade of adjacent sea weeds. It assumes a
variety of shapes ; it can change its locality at pleasure ;
and, having apparently an aversion to the solitary life, it is
usually found in clusters of perhaps a dozen or more in-
dividuals huddled together in close compact. The species
can scarcely be mistaken. In the same pool are other
species of the same genus. The most abundant of all the
anemones — Actinia mesembryantJiemum — is seen on all sides.
It inhabits nearly every place at the sea side, adhering to
stones or to the sides of piers, and is easily recognized, either
as a gelatinous rounded body, not unlike a ripe cherry, when
closed during the recess of the tide, or, when in the full
enjoyment of the returning waters, it exhibits the form of
the flower from which its name has been derived. A care-
ful examination of some of the deeper pools in this neigh-
bourhood will probably expose to view very fine specimens of
the larger species — Actinia coriacea and Actinia crassicornis.
The former of these is abundant on many parts of the shore,
sometimes buried in the sand, and gregarious, or hidden
under fragments of rock between the tide marks, or in the
possession of a luxuriant pool. It may be known by the
following characters : — Body conoid, or wider at the base
than above ; skin opaque, coriaceous or leathery, covered
with warts, variously coloured, but usually dull red ; tenta-
cula numerous, in three or four series, aud, when expanded,

25G MARINE ZOOLOGY OF ARRAN.

marked with rings. The other species, A. crassicornis,
although commonly found attached to shells or stones in
deep water, may often be obtained in rock crevices, at low-
water during the spring tides, or even in the large pools. It
may be prociired at Corriegills, or among the rocks at the
south-east side of Holy Island. It resembles A. coriacea in
shape ; but it is a finer species ; it grows to a larger size, it
is less coriaceous, and more vivid in its colours. It is a
handsome object for the aqua-vivarium, where, after a short
period, it will be observed that the body becomes filled with
water, so as to be greatly enlarged and diaphanous, or almost
transparent ; the tentacula, at the same time, are elongated,
increase their diameter, and exhibit their tubular form.
This diaphanous appearance is perhaps the most striking
character by which it is distinguished from A. ccn'iacea. It
will add considerably to the interest of the vivarium — pro-
vided the tank can afford accommodation — if three or four
individuals of this fine species are placed artistically among
the rock-work ; for, by a careful selection, specimens may be
introduced of variegated hues, — some of cream colour, some of
scarlet, some of saffron, with orange stripes, or marbled with
red and white. There are, indeed, certain people who enter-
tain other ideas as to the proper destination of Actinia
crassicornis. "For, of all kinds of sea anemones," says
Dicquemare,* " I would prefer this for the table ; — being
boiled some time in sea water, they acquire a firm and palat-
able consistence, and may then be eaten with any kind of
sauce. They are of an inviting appearance, of a light
shivering texture, and of a soft white and reddish hue.
Their smell is not unlike that of a warm crab or lobster."
But whatever importance the epicure may attach to the above
information, the sea-side visitor will do well to watch the
manoeuvres by which the anemones are themselves enabled to
gratify their tastes by a variety of dishes. It is most amus-
ing to observe the apparent cleverness — the agility and
* Johnston's British Zoophytes, vol. i. p. 227.

MARINE ZOOLOGY OF ARRAN. 257

artifice exercised by these curious creatures for the capture of
their food. " On one occasion," says Mr. Couch, " while
watching a specimen (A. coriacea) that was covered merely by
a rim of water, a bee, wandering near, darted through the
water to the mouth of the animal, evidently mistaking the
creature for a flower, and, though it struggled a great deal
to get free, was retained, until it was drowned, and was
then swallowed." * They are at once the most abstemious
and the most gormandizing of animated beings. They will
live without food for upwards of a year, and yet they may
be seen at all hours, and every day, angling, as it were, with
their tentacula, and catching crabs, pi-awns, limpets, peri-
winkles, dog-whelks, small fish, and, in short, whatever, in
the shape of fish, flesh, or fowl, is brought within their
reach. They retain their food for ten or twelve hours, and
then eject from the stomach the well-picked bones — the
emptied shells of the crustaceans and mollnsca. Occasionally
a bone will stick in the throat. " I had once brought me a
specimen of A. crassicornis that might have been originally
two inches in diameter, and that had somehow contrived to
swallow a valve of Pecten maximus of the size of an ordinaiy
saucer. The shell, fixed within the stomach, was so placed
as to divide it completely into two halves, so that the body,
stretched tensely over, had become thin and flattened like a
pancake. All communication between the inferior portion
of the stomach and the mouth was of course prevented ; yet,
instead of emaciating and dying of an atrophy, the animal
had availed itself of what undoubtedly had been a very
untoward accident, to increase its enjoyments and its chances
of double fare. A new mouth, furnished with two rows of
numerous tentacula, was opened up on what had been the
base, and led to the under stomach ; — the individual had in-
deed become a sort of Siamese twin, but with greater
intimacy and extent in its unions." t Another remarkable
character in the natural history of the Actinias is, that,

* Johnston's British Zoophytes, vol. i. p. 225. "fr Ibid. p. 235.

s

258 MARINE ZOOLOGY OP ARBAX.

though impatient of ill-treatment under certain conditions,
as when torn abruptly from their attachment to the rock, or
when confined in water not sufficiently pure, they are almost
indestructible by the usual methods of destruction. " They
may be immersed in water, hot enough," says Dr. Johnston,
" to blister their skin, or frozen in a mass of ice and again
thawed; and they may be placed within the exhausted
receiver of the air-pump, without being deprived of life, or
disabled from resuming their usual functions when placed in
a favourable situation. If the tentacula are clipped off they
soon begin to bud anew, and if again cut away they grow
again." * The finer specimens of these two species, as found
in this locality, will occasionally measure from four to six
inches across.

112. "With the exception of the above examples, the coast
line of Arran is not prolific in a variety of zoophytes. A few
other forms may be procured in deep water by means of the
naturalist's dredge. There are, however, two more species
that will possibly attract the attention of the inquisitive
rambler along the shore : they belong to genera closely allied
to Actiniw — namely, Anthea cereus, and Adamsia palliata.
The former frequently chooses for its position the leaves of
the grass wrack, or sea grass (Zostera marina), extensive beds
of which constitute submarine meadows in sandy districts.
This wrack (including the other species, Z. nana) is the only
instance of a British flowering plant that lives in the sea.
Now, wherever there is a bed of this plant, the student of
marine zoology (and also the algologist, or marine botanist)
may pause and examine ; and if he does not object to wade
knee-deep, at ebb-tide, he will almost to a certainty be repaid
for his trouble. Large numbers of creatures revel and thrive
in this forest — fish, crabs, shell-fish, annelids or sea-worms,
and zoophytes ; and here, dependent from the long ribband-
like leaves of the zostera, you may often discover Ant/tea
cereus, with his lengthened tentacula, on the look-out for his

* Johnston's British Zoophytes, vol. i. p. 239.

MARINE ZOOLOGY OF ARRAN. 259

prey. The following characters will be sufficient for iden-
tification : — Body cylindrical, smooth, adhering by a broad
base; tentacles numerous, longer than the body, and unlike
those of the Actiniae, these arms cannot be retracted, or are
scarcely retractable, into the body of the animal. It attains
in this neighbourhood the size of about three inches, measured
from the base to the oral disc, or mouth ; and it is of a light
brown, or dull ash colour. Beds of the Zostera marina occur
at Lainlash, on the sandy shore opposite the village; also, on
Holy Island, not far from St. Molios' Cave; but, even where
the sea wrack does not grow, AntJiea cereus may be found
attached to one of the commonest of our sea-weeds — Fucus
serratus. The other species — Adamsia palliata — will not so
readily be discovered by the casual visitor, for it prefers the
deep water. The dredge, when let down to ten or twenty
fathoms, will often bring up a dozen specimens at a time.
This zoophyte may be grouped amongst the most curious of
creatures. It will be sure, when once seen, to arrest attention.
Its habitat, or dwelling-place, is the exterior wall of a deserted
shell, as, for example, the dead shell of a buccinum, or whelk,
or of a trochus, or fusus, over the greater part of which the
animal is extended as a flattened mass about three-tenths of
an inch in thickness, varying from a light brown to cream
colour, and having the whole surface of its body streaked
and spotted. The situation of the oral disc is distinguished
by three or four rows of tentacula, short and white, forming
an oval margin round the mouth. A thin substance, like
horn, will be observed covering part of the empty shell, to
which horny material the body of the Adamsia is attached ;
and, when excited, apparently by pleasure, as with feeding,
it emits long filaments, like white threads or delicate silk
cords. But a singular circumstance has yet to be men-
tioned:— The dead shell, over which the zoophyte is spread,
is generally inhabited by a hermit crab, and always, as is
supposed, by the same species, namely Pagurus Prideauxii.
It is exceedingly curious to watch how advantageously to

2GO MARINE ZOOLOGY OF ARRAX.

both parties the arrangement works. Of course the Adamsia,
without any fatigue or effort, is carried by the roaming
propensities of the crab over a large district; and, in this
way, he commands an extensive market for the acquisition
of food; whereas the Actiniae, being fixed to rocks or half
buried in the sands, must either undertake a slow and
wearisome journey, by their own unassisted labour, or be
satisfied with the supplies brought by the wind or tide
within reach of their feelers. The hermit, on the other hand,
is also recompensed ; for, as the writer of these notices has
frequently observed, the palatable morsels secured by the
tentacles of the zoophyte are instantly seized by the claws of
his crustacean companion, and, without any apparent apology
or subsequent remorse, are partly appropriated to his own
immediate use. " In all likelihood," says the late Rev. Dr.
Landsborough, " they in various ways aid each other. The
hermit has strong claws, and while he is feasting on the prey
he has caught, many spare crumbs may fall to the share of
his gentle-looking companion. But soft and gentle-looking
though the Actinia be, she has a hundred hands, and woe
to the wandering wight who comes within the reach of one
of them ; for all the others are instantly brought to its aid,
and the hermit may soon find that he is more than com-
pensated for the crumbs that fell from his own booty." *
Specimens of this curious and beautiful zoophyte may occa-
sionally be procured by a search in the zostera beds at
Lamlash, or in other places at low water. Dr. Lands-
borough first saw it at the mouth of the Glen Rosa burn, in
Brodick Bay.

113. We must now inquire about the prawn captured at
Corriegills. Look at it through the sides of the glass vessel.
The creature is almost transparent. This is Palcemon
gquUla; it is smaller than the common prawn, Palcemon
serratus, and differs in a few other points from that species.
It is readily found in the rock pools on Holy Island, on both

* Landsborough, Popular History of British Zoophytes, p. 230.

MARINE ZOOLOGY OF ARRAIf. 26

sides of Brodick Bay, and indeed almost everywhere on the
Arran shore, and on the Ayrshire coast. It is an interesting
object for the vivarium, both on account of its general appear-
ance and from its activity in plundering the sea anemones.
Other crustaceans abound in this district. The numerous
rocks are places of retreat and shelter for the common edible
crab, Cancer Pagurus, which, in moderate size, is caught in
the creels by local fishermen, whilst the younger individuals
of this species may be recognized scampering about the shore
in search of food, or hiding themselves under stones, in the
enjoyment of a siesta. The common shore or harbour crab,
Carcinus Mcenas, inhabits the same localities, and may be
seen anywhere, or everywhere, in pools, under sea-weeds,
beneath rocks, or in the sands. Turn over the stones, and
another species will invite inspection — Porcdlana platychdes
— distinguished at once from every other crab by the breadth
of its hands. But, of all the crustaceans, the hermit crab,
already mentioned in connection with an associated zoophyte,
seems to afford the most amusement to the sea-side rambler.
There are several species of this singular crab inhabiting deep
water; but the most common kind, Pagurus Sernhardus,
may be procured in any quantity by strolling along the
shore. The peculiarities of these hermits are of no mean
order. Each individual resides, hermit-like, solitary in a
shell, which, either by fair or foul means, is obtained and
appropriated for self-protection. Nature has been considerate
in supplying a shield (the carapace) that entirely defends
every species of crab, except the genus now under our notice,
namely, Pagurus, which, whilst tolerably safe as to the head
and thorax, is in constant danger from the utterly unpro-
tected state of the posterior portions. The abdomen and
tail are without the usual covering of a coat of lime ; they
have no sJtell in these parts, and, consequently, at any
moment they might be at the mercy of a hungry neighbour
or spiteful antagonist. But here, as elsewhere in the wonder-
ful economy of Nature, the animal finds compensation for the

2G2 MARINE ZOOLOGY OF ARRAN.

deficiency, either in the use of its instinct, or rather, perhaps,
in the exercise of good sense, by which a remedy is provided
equal to its necessities. Its usual habit is to take possession
of the empty shell of a dead shell-fish — one of the univalves,
such as the common whelk — and, introducing the whole of
its soft parts into this cherished prize, it hooks its tail round
the innermost whorl of the shell, and keeping both eyes and
claws ready for action, it bids a hearty defiance to every
opponent. You may seize the exposed claws and tear the
hermit into pieces; but, so apprehensive are they of danger,
and so tenaciously do they cling to their support, you will
seldom either persuade or force the creatures to come out of
their abodes. It is equally curious to observe the cleverness
shewn when required to meet another difficulty. A young
hermit occupies a small house. This is a necessary con-
clusion, inasmuch as the shell has to be dragged about by
its occupant. But juvenile hermit crabs, like other young
creatures, increase in size, and, as they grow, they find that
the abodes selected in their youth have become incon-
veniently small; for, of course, the dead shells cannot be
enlarged for their accommodation. This embarrassment is
easily obviated. It is merely a matter of change of residence.
You will see the hermit — if you watch on the sea shore, and
happen to be in the right place at the right time — bring
himself into close proximity to a shell, empty, and larger
than the one then in his possession. His next step is to
take an accurate survey on all sides, to see that the coast is
clear, and to make sure that he can move from one house to
the other without the risk of a disadvantageous encounter
with an enemy. It is the work of a moment, provided that
the new house proves to be more commodious than the old
one; but it sometimes happens that the crab is mistaken in
his calculation, and, in this case, it is most amusing to watch
our friend running about, half naked, and half frightened,
among a group of deserted shells, poking its tail first into
one and then into another, twisting itself in various direc-

MARINE ZOOLOGY OF ARRAN. 263

tions, and assuming different attitudes, in order to test the
suitability of the proposed arrangement; and, finally, when
fitted to its entire satisfaction, it walks off, evidently pleased
with its success, and generously leaves the old shell for the
accommodation of any future house-hunter less corpulent than
itself. Rot unfrequently two hermits will meet and dispute
the possession of a vacant shell. Arbitration is never sought
in these cases. It is simply a contest between the two for
» mastery — a trial of physical power ; and, where the com-
batants are well matched, the conflict is generally of a fatal
character ; for the possession of a convenient house is con-
sidered of such vital importance that the one or the other is
determined to obtain the coveted shell, and, at all times
pugnacious, each is resolved to die rather than yield in-
gloriously to its opponent. This sea-fight often occurs.
The slain are not a few. Sometimes a less serious encounter
may be witnessed, as when mig/it exercises its too frequent
tyranny over riyJd. The following incident affords a good
illustration : — Two hermits met accidentally, it may be sup-
posed, within the vivarium on Holy Island. The one was
somewhat stouter than the other; and the weaker brother,
appearing at first sight to possess the larger house, received
an immediate summons to turn out and vacate his premises.
The summons was disputed, a furious combat commenced,
and, after a variety of manoeuvres, with thrusts and counter-
thrusts, the stronger, with a dexterous use of the hands,
seized the little fellow by both his wrists, and endeavoured
to shake him out, despite the opposition that was offered.
The shells rattled against the glass wall. It was a fierce
engagement. At last the older hermit, coming partly out of
his shell, seemingly for the purpose of obtaining a better
purchase and a greater command of his strength, pulled, and
pulled, and pulled again — still drawing himself farther out
of his shell, as if he felt that the work was desperately tough —
until the youngster giving way, stood, unhoused, discomfited,
and embarrassed, to abide the good pleasure of his master.

2G4 MARINE ZOOLOGY OF ARRAN.

The next scene was amusing enough. The conqueror came
entirely out of his house, and — retaining a firm grasp of his
victim to prevent him from bolting off with either of the
two shells now empty — backed himself into the new abode,
just to inquire if it would really prove a more convenient
residence than his own ; but, disappointed in his anticipations,
he returned to his former shell, released his young friend
without further detention or injury, and apparently, with a
polite " good morning," they separated, as if nothing dis-
courteous, or at least unusual, had occurred.

114. The star-fishes will also afford interest during a
ramble along the shore. The most common species, Uraster
rubens, is really too common to be further noticed. Some
kinds, as Palmipes membrcmaceus, Cribella rosea, Asterias
aurantiaca, and others, can be procured only by the dredge;
but several of the deep water species may occasionally be
found by wading through the zostera beds at Lamlash, or by
examining the pools and rock crevices, especially after a
storm, on Holy Island. In this way, Luidia fragilissima,
Uraster glacialis, Cribella oculata, Goniaster Templetoni,
Solaster papposa, with some of the species of Ophiocoma and
Ophiura, may be picked up without much trouble. In a few
places, the smallest of British star-fishes, Asterina gibbosa,
may be obtained; but good eyes and some patience are
required. Try the pools at the north end of Holy Isle ; and
also at Clachland Point, opposite. Turn aside the sea-weed;
and it may be, after considerable search and disappointment
— for this species is by no means abundant — you will see the
little creature, grayish in colour, and gibbose and augulated
in form, adhering to the side of the rock. In Lamlash Bay,
Uraster ylacialis is plentiful. It may often be observed on
a calm day, when the water is as clear as crystal, directing
its course among the algae at the depth of ten or fifteen feet
— the creature itself, perhaps, two feet in length — and, on
nearer inspection, you will distinguish it at once from
Uraster rubens by the spines situated on the back of each

MARINE ZOOLOGY OF ARRAN. 265

ray. Care is needed in the preservation of this species, for
it has the knack of dislocating its arms under your most
cautious treatment. But the star-fish that has gained most
notoriety, from the facility with which it can apparently at
pleasure dismember itself — nay, break itself into any number
of fragments — and, vexatiously enough, at the very moment
when you think the animal is deceased, and preserved un-
injured, is Luidia fragilissima. The subjoined description is
from the pen of the late Professor Edward Forbes : — " The
first time I ever took one of these creatures, I succeeded in
getting it into the boat entire. Never having seen one
before, and quite unconscious of its suicidal powers, I spread
it out on a rowing bench, the better to admire its form and
colours. On attempting to remove it for preservation, to
my horror and disappointment I found only an assemblage of
rejected members. My conservative endeavours were all.
neutralized by its destructive exertions; and it is now badly
represented in my cabinet by an armless disc and a discless
arm. Next time I went to dredge on the same spot, deter-
mined not to be cheated out of a specimen in such a way a
second time, I brought with me a bucket of cold fresh water,
to which article star-fishes have a great antipathy. As I
expected, a Luidia came up in the dredge — a most gorgeous
specimen. As it does not generally break up before it is
raised above the surface of the sea, cautiously and anxiously
I sunk my bucket to a level with the dredge's mouth, and
proceeded in the most gentle manner to introduce Luidia to
the purer element. Whether the cold air was too much for
him, or the sight of the bucket too terrific, I know not, but
in a moment he proceeded to dissolve his corporation, and at
every mesh of the dredge his fragments were seen escaping.
In despair I grasped at the largest, and brought up the ex-
tremity of an arm with its terminating eye, the spinous
eyelid of which opened and closed with something exceedingly
like a wink of derision." *

* British Star-fake*, p. 138.

266 MARINE ZOOLOGY OF ARRAN.

Experience has since shewn that the means by which either
Luidia or Uraster glacialis can be secured for cabinet speci-
mens is, not to destroy them violently by plunging them into
fresh water, but to let them die by a more quiet process —
viz., by leaving them in a vessel of sea water until life is
exhausted. Under these circumstances, they are not prone
to break, and thereby disappoint their captors.* Luidia will
be recognized by its light orange or buff colour; its rays are
smooth on the back, and provided with spines on their mar-
gins ; specimens measuring a foot may be picked up on the
shore in Lamlash Bay, and individuals of twice that size are
procurable by the dredge, or may sometimes be hooked up in
shallower water with a common rake. There is another
singular creature of great interest to the naturalist — it is
Comatula rosacea, or the rosy-feathered star-fish — a beautiful
and elegant example of the radiated form of animated beings.
It is found abundantly in Lamlash Bay, near the pier on
Holy Island, and in other parts of the locality; but the
dredge is required, as this species inhabits deep water — that
is, depths of about ten fathoms. At some seasons almost any
number may be obtained. They are brought up attached to
the large sea-weed, Laminaria saccfiarina, from which they
must be removed with care, as they are exceedingly brittle,
and, like the Luidice, can break themselves into fragments
with astonishing and unpleasant speed. The best way, per-
haps, to prepare dry specimens for the cabinet is to treat
them as the marine botanist treats the more delicate algse —
spread them on drawing paper, place over them a piece of
smooth linen, and let them dry between folds of blotting
paper, under slight pressure; but, in the first instance, they
must be allowed quietly to die in sea water, or more rapidly
in a solution of alum — for, if immersed in pure fresh water,
they will lose their beauty, by the removal of their colouring
matter. In order to appreciate the exquisite form and ele-

* Dr. Carpenter has found that, by placing Lvidia in a pan of
Glyc'-rine, the creature dies at once, and without disfiguring itself.

MARINE ZOOLOGY OF ARRAN. 26 7

gant movements of this star-fish, the creature must be seen
and watched in a vivarium, where it will voluntarily fix itself
to the sea-weed or to a piece of rock, and, by graceful undula-
tions of its arms, will be certain to command admiration.
The picture will be greatly enhanced if several individuals of
various colours — orange, purple, crimson — ai-e introduced
and judiciously dispersed.

115. Before we leave this division of the subject, it will
suffice to make brief mention of the well-knowu sea hedge-
hog, Echinus sphcera, common in most parts of the Clyde, and
of which fine specimens are found clinging to the rocks on
Holy Island, and at Clachland Point. This curious creature
should be carefully examined by the young student of marine
zoology. It is allied to the star-fishes ; for, although spherical
in form, the radiated structure is readily perceived. In
shoi-t, it is a star-fish, with the spaces between its rays filled
up by plates of carbonate of lime — the rays themselves con-
sisting of the same material — the whole exterior being bent
over into a hollow ball, and armed, hedgehog-like, or like the
star-fish Uraster glacialis, with numerous sharp spines. The
viscera, or digestive organs, of the Echinus are contained
within the ball ; and its mouth is provided with a beautiful
piece of mechanism, worthy of examination, and designed for
crushing the shells of molluscous and crustaceous animals on
which it feeds. Its mode of progression, by means of its
spines and suckers, is both interesting and wonderful. Speci-
mens are frequently found of the richest crimson or purple.
Another species of this genus, Echinus miliaris, is also common
in the pools. It is more diminutive than the former. The
dredge will generally be needed to obtain living specimens
of the other forms of these echinodermatous or radiated
creatures — such as Spatangtis purpureus, Echinocyamus
pusillus, and Amphidotus cordatus, or the common Heart
Urchin, of which the dead and empty shells, with their spines
rubbed off, may frequently be noticed washed up and left
upon the sands.

268 MARINE ZOOLOGY OF ARRAN.

11C. The study of another important division of animated
nature — the mollusca — has been a favourite pursuit of scien-
tific men and of amateur observers. To this division the
marine shells belong. Here the conchologist finds his delight
in the examination and arrangement of the materials of his
department — admiring the endless variety of form, and colour,
and sculpture; whilst the zoologist is laboriously occupied in
determining the anatomical structure of the inhabitants of the
shells — tracing the peculiarities of the different genera and
species.

The waters around Arran produce a considerable number
of shells; but, with few exceptions, the species are common
to all parts of the Clyde. The dredge is required to procure
specimens of interest ; and, for the information of naturalists
unacquainted with the district, it may be stated that the
most profitable dredging ground, in Lamlash Bay, lies between
Hamilton Rock, near Clachland Point, on the Arran shore,
and the north and north-east sides of Holy Island, extending
the whole way across, and in depths from fifteen to forty
fathoms. A full list of these species will be given at the end
of these notices ; but the following genera, as being found in
this particular locality, may be here mentioned: — Aporrhais,
Artemis, Astarte, Cardium, Cerithium, Ghemnitzia, Circe,
Corbvla, Crania, Cylichna, Cyprtea, Cyprina, Dentaliui/i,
Emarginula, Eulima, Fissurella, Fusus, Kettia, Leda, Lima,
Lyonsia, Afangelia, Modiola, Montacuta, Nassa, Natica,
Nucula, Odoslomia, Pecten (including P. striatus and P.
tigrinus), Pectunculus, Philine, Pileopsis, Pilidum, Psam-
mobia, Puncturella, ftissoa, Scalaria, Scaphander, Tellina,
Terebratula, Thracia, Trichotropis, Trochus, Tttrritella, and
Venus. Several of these genera occur also in other places in
Lamlash Bay, and round the coast. On the other hand,
Brodick Bay, Whiting Bay, and the vicinity of Pladda, have
hitherto proved to be exceedingly unprofitable. Loch Ranza,
notwithstanding the apparent advantage of its position, did
not yield anything, after several hours of active dredging in

MARINE ZOOLOGY OF ARRAX. 2G9

all parts of the bay, except the commonest of the scallops,
Pecten opercularis — an excellent bait for fishermen, and which
may be procured here in any quantity. The strait which
separates Arran from Argyleshire, known as Kilbrannan
Sound, will probably yield richer results than other parts of
the Clyde; but these waters have not yet been sufficiently
examined to warrant more than a conjectural opinion. The
south end, and the south-east side, of Holy Island, are also
unprofitable.

It must not be expected that the rarer molluscous animals
or shells may be obtained from the pools, or from the sands,
or rocks forming the coast-line ; yet an attentive investiga-
tion will not be without its reward. In the vicinity of Clach-
land Point, where the excavated sandstone affords numerous
places of retreat, and where each returning tide supplies the
wants of the tenants, a variety of creatures may be seen
grouped together in quest of the enjoyments of life. Let the
crevices be carefully searched — turn over the fronds of the
sea-weeds — capsize the loose stones — look under the ledges of
the rocks — select a calm day, during the spring tide, when
the water has ebbed to the lowest — and there will not be any
need, at least for the young student, to complain of the result.
A curious mollusc inhabits this station — Aplysia hybrida, or
the sea hare, which will be recognized, whilst in a state of
activity, by the peculiar shape of its antennae; for these, when
expanded, are something like the erect ears of the common
hare. Another mollusc found upon the shore, and not with-
out pretensions to respect, as well from its lovely orange
colour as from the gracefulness with which it moves, is the
Pleurobrancfms, which, being nocturnal in its habits, may be
caught napping at the period most convenient to its captor.
It is a good object for the vivarium, where, lying concealed
during the day, it will be observed, at midnight, traversing
the tank like a solitary watchman on his rounds; but, unlike
the guardian of our streets, its purpose is — not to defend pro-
perty, but to appropriate to its own use, without fear of

270 MARINE ZOOLOGY OF ARRAN.

detection, whatever it may regard as palatable to its taste or
requisite for its necessities. Both the Pleurobranchus and
the Aplysia may be procured from pools at the north end of
Holy Island. At Lamlash, in the neighbourhood of the old
quay, the blocks of sandstone, which lie scattered in all direc-
tions, contain here and there good specimens of one of the
borers — a division of the mollusca, whose habit is to penetrate
either sand — as the common cockle, Cardium edule — or wood,
as the different species of Teredo — or stone, as the genus
Phola*, a bivalve, of which the shell in some species is ex-
tremely fragile, and although wonderfully well adapted to
its work as an excavator of solid rocks, requires the utmost
caution when being handled in our collections. The species
that bores through the sandstones at Lamlash is Pholas
crispata. The wood-borer, Teredo navalis, is common in
many parts of the Clyde, as may be proved by witnessing its
destructive operations at Ardrossan and Fairlie, on the Ayr-
shire side, where also its ally, Xylophaga dorsodis, has done
its full share of mischief Another species, less common,
and of larger growth — formerly unknown, as is supposed,
higher up the Clyde than Port- Patrick — namely, Teredo Nor-
vegica — introduced itself into Lamlash Bay, and, during the
short period of about seven years, it had almost demolished
the massive supports of the pier — a commodious and sub-
stantial landing-place, erected, as the owner undoubtedly
thought, for the convenience of himself and friends. The
Teredo, however, commenced and continued its operations
unnoticed — not oae pile of the water-covered timber escaped
— the whole was pierced in lengths varying from a few inches
to about two feet, when, in the hurricane of February, 1856,
a vessel was driven upon the spot, and almost the entire
structure was swept away.* On an examination of the frag-
ments thrown upon the island, several specimens of the cal-
careous tube, formed by the animal in the course of its pro-

* This pier was built by the late Mr. Oswald, member for the city
of Glasgow.

MARINE ZOOLOGY OF ARRAX. 271

gress, and a few of the valves, were secured for private
cabinets; and pieces of the bored timber were given to the
museums of Glasgow College and of the Andersonian Uni-
versity.

The most abundant species of Trochus on these western
shores is T. umbilicatus, which, with T. cinerarius and T.
tumidus, may be procured in any quantity on Holy Island.
Here also the following shells are not uncommon: — Acmcea
testudinalis, Kellia rubra — a minute bivalve attached to the
dark low growing plant, Lichina pygmcea, which overspreads
the rocks near the sea; Patella athletica, P. pellucida, Tapes
decussata, with different species of Littorina, and other equally
common and widely diffused forms. In the sands, at Lam-
lash, at low water, Mactra sollda and M. subtruncata are
plentiful. But, leaving the coast line, and letting the dredge
drop in about ten fathoms, at the north end of Holy Island,
or between the north point of the island and the pier, a
peculiarly interesting shell may be collected in large quan-
tities— Lima hians — the inhabitant of which constructs and
occupies a nest. The shell is of delicate texture, and, when
deprived of extraneous matter by careful washing, is entirely
white, and, being graceful in form, is an acquisition to an
ornamental cabinet. It is desirable, however, to preserve
specimens of the nest as well as of the shell, and to allow
both to remain, as far as possible, in their natural position.
The nest is formed of materials collected at the bottom of the
sea — either pebbles, or broken shells, or both shells and
pebbles; and these are brought into a mass and bound
together by a glutinous thread secreted by the animal.
Sometimes the Lima is solitary — having built a house simply
for itself ; but, generally in Lamlash Bay, the dredge brings
\ip large patches of this compacted debris, in which will be
found a colony living in apparent comfort, security, and
friendship — each individual, however, having a separate and
snug berth. A supply of specimens may be secured in a few
moments. The animal is also a 'beautiful object in the water

2i'l MARINE ZOOLOGY OF ARRAJT.

— its numerous tentacula, which extend beyond the limits of
its shell, are of a fine orange colour, and, being a bivalve, it
moves from place to place with ease and agility, by the rapid
opening and closing of its valves. Although rare in many
localities, Lima hians has an extensive range in the Clyde —
the nests being found off the coast of Islay, between Largs
and the Cuinbraes, and on the Argyleshire side, as high up
as Hunter's Quay, Dunoon. But the station where it is
really abundant is Lamlash Bay, where also the other species,
L. loscombii, and L. subauriculata, may be procured.

117. Although many objects of interest have now been
mentioned, there remain a considerable number of living
creatures, more or less wonderful both in structure and in
habit, which, from various causes, have not yet been sufii-
ciently examined on this coast; and, consequently, any in-
formation respecting them must necessarily be meagre.
Among the Annelids, or sea-worms, the following genera are
known to belong to Lamlash Bay: — Aphrodita, Arenicola,
Eunice, Nereis, Pectinaria, Serpula, Spirorbis, Terebella.
The young student will be usefully employed in the search
and study of these curious forms of life ; for, notwithstanding
that they take no higher rank than worms, they exhibit, in
some species, characteristics so peculiar, and colours so gor-
geous, as to attract even the popular eye, and to excite un-
sparing admiration. Who is not familiar with the Serpulce,
whose heads are crowned with radiating threads of varied
hue ? These singular tufts, which the animal protrudes for
health and pleasure, and which, with the rapidity of thought,
are withdrawn into its calcareous tube on the first symptom
of alarm, are its gills, or organs of respiration. Let the
shadow of your hand pass near the side of a glass vessel, in
which a living specimen is contained, and instantly the head
starts back into concealment; but watch, and in a few
moments its brilliant coronet will re-appear. The Pectinaria
resides in a house made of the finest sand, cemented in the
form of a tapering tube; and its gills, in shape like a minute

MARINE ZOOLOGY OF ARRAN. 273

comb, are as bright as burnished gold. In the Terebellce,
inhabitants of mud, the numerous and long worm-like ten-
tacles present the appearance of so many separate Annelids
entwined around their common prey. The genus Eunice
occupies a tube composed of a substance not unlike thin horn,
or the slender quill of a bird. The spiral white spots, fre-
quently spread over tangle or other large sea-weeds, are
examples of Spirorbis. The sea-mouse, Aphrodita aculeata,
obtained with the dredge, will easily be recognized by the
metallic lustre of its long bristles, which, partially covering
the animal, give out the colours of the rainbow. Some of
the marine worms are remarkable from the enormous length
to which their thread-like bodies are extended. Another
class of widely-diffused beings, not yet sufficiently investi-
gated on the Arran shores, and which, like the Annelids, are
worthy of a more honourable name than is assigned to them
in popular language, are — the Sea-slugs, known in scientific
phraseology as the Nudibranchiate mollusca. These are
within the reach of every observer ; for, at low water, they
may be seen reposing under loose stones, or adhering to Algce.
Some of the species, especially of the genus Eolis, are really
charming objects — elegant in form and beauteous in colour.
They cannot be mistaken. Place a specimen in a tumbler of
sea- water, it will unfold itself, and its gills, differently situated
in different species, will be seen expanding into full operation
as the little creature pursues its journey round the sides of
the vessel. It enjoys the learned appellation of Nudibranch,
because its branchiae, or breathing organs, are naked, or ex-
terior to the body of the animal. The genera Doris, Eolis,
Goniodoris, Lomanotus, Polycera, and Triopa, have been
found in the bays of Brodick and Lamlash ; but, if an active
search be made, the number of ascertained species will doubt-
less be greatly augmented, and the labourers in this depart-
ment will be amply rewarded. Nor must we altogether
overlook, as among the more remarkable forms, the Acalepha,
to which the Medusae and other Jelly-fish or Sea-nettles be-

T

274 MARINE ZOOLOGY OF ARRAV.

long. Some of the larger and coarser species are familiar to
sea-bathers by the stinging qualities of their tentacula. But
there are other species as harmless as they are lovely. On a
calm summer day, when there is not a ripple to disturb the
sea, these exquisitely formed creatures may be witnessed in
hundreds, like a vast fleet of faiiy ships, lying upon the sur-
face. Such are the Beroe and the Cydippe. Let specimens,
captured carefully in a gauze net, be transferred to a vessel
of water, and, low as they are in organization, they will not
be dismissed without commanding surprise at the delicacy of
their structure, and at the facility with which they traverse
their allotted space.

118. It remains only to add, after the above general view
of invertebrate life, that in the various bays and inlets around
Arran almost every kind of fish common to our northern
seas may be procured. It would be superfluous to name the
species familiar to every person. But we must not omit to
mention that the Lancelot — Amphioxus lanceolatus — has been
dredged at the north end of Holy Island, where it seems to
be restricted to a gravel bottom in depths of about ten or
fifteen fathoms. Until recently this fish was regarded as
extremely rare. It is now, however, known to be more
common. Several specimens have been captured near Mill-
]x>rt, in the island of Cumbrae, as well as in Lamlash Bay ;
and probably it will be discovered to have a wider range
than at one time was expected. The interest attached to it
arises from its anatomical and physiological peculiarities.
Its spine is a cartilaginous, thread-like column, without
joints ; it has no ribs, no pectoral or ventral fins ; and, in
short, although claiming rank with vertebrate animals, the
skeleton is rudimentary, and the brain absent. The
A mphioxus is an excellent illustration of the law that, even
when there is the greatest departure from uniformity, the
typical characters are rigidly preserved in the development of
creation.

119. In concluding our notices of the marine zoology of

MARINE ZOOLOGY OF ARRAN. 275

Arran, the subjoined tables may advantageously be added
for the information of naturalists not practically acquainted
with the Fauna of the Clyde. The species marked with an
asterisk were dredged by the late Rev. Dr. Landsborough and
Major Martin; and the other species were obtained by Dr.
Greville and myself, in our examination of this part of the
coast. The list, so far as completed without the assistance of
the former gentlemen, appears in the "Annual Report of the
British Association for the Advancement of Science for the
year 1856." Dr. Landsborough and Major Martin were inde-
fatigable in their investigation of these localities. There yet
remains, however, much work to be accomplished by future
inquirers — especially among some of the remarkable groups,
as the Cirripeda Annelida, Acalepha, Zoophyta, and
Poriphora — as also among the Polyzoa and microscopic
forms included in the Infusiora and Rhizopoda.

NOTE BY WILLIAM B. CARPENTER, M.D., F.R.S., <fcc.

120. I am not able to add anything to Dr. Miles' list except
on one point, which is a very curious one. There are two species
of Comatula (Art. 109) in Lamlash Bay; one of them, the
commonest, being that described by Edward Forbes as
Comatula rosacea; whilst the other, rare there, but tJte
species, of the south of England, is that described by Miller
in his " Crinoidea," and designated by Miiller, Comatula
Milleri. I have reason to think that the latter — not the
former, as believed by Edward Forbes — is identical with the
Comatula Mediterranea of Lamarck. But this I cannot take
upon myself certainly to say, until I have had a specimen of
the C. Mediterranea for careful examination. Putting aside
other differences, the distinction between the two species of
Lamlash Bay is well marked by the presence in C. Milleri of
what Miller has described and figured as a single inter-radial
plate in each of the angles between the origin of the rays,

276

MARINE ZOOLOGY OF ARRAN.

but which is really formed of three minute plates, which I
suspect not to be inter-radials at all, but dermal plates; of
these there is no trace vrfiatever in the Comatula rosacea.

TABLE I. — MOLLUSCA.

*Aclis nitidissima, .
* unica, .

Acmaea testudinalis,

virginea,

*Adeorbis sub-carinata,
*Amphysphyra hyalina,

Anomia ephippium,

aculeata,

Aplysia hybrida,

Aporrhais pes-pelicani,
*Arca lactea, .

Artemis exoleta,

lincta,

Astarte compressa,

* elliptica,

Scotica,

— sulcata,

Buccinum undatum,
*Co3cum glabrum, .
— trachea,

Cardium echinatum,

edule,

* iasciatum. .

* nodosum, . . »

Norvegicum,

* Suecicum,

*Cerithiopsis tubercularis,
*Cerithium adversum,

reticulatum, .

Chiton asellus,

marmoreus, .

ruber, .

*Chemnitzia indistincta, .
* rufescens,

Circe minima,

The whole of the shells named
in this table were dredged
in Lamlash Bay — the best
ground being the area that
extends from Clachland
Point and Hamilton Rock
to the north end of Holy
Island, in from 15 to 40
fathoms.

In the pools, Clachland Point.

Scarce in Lamlash Bay.

MARINE ZOOLOGY OF ARRAN.

277

Corbula nucleus,
Crania anomala,

*Crenella decussata,

discors,

* marmorata,

Cylichna cylindracea,

* mammillata,

* obtusa,

* truncata, .

umbilicata,

( Attached to stones and shells
< in deep water — not scarce
( in Lamlash Bay.

Cypraea Europaea, .

Cyprina Islandica, .

Dentalium entalis, .
*Donax anatinus,

Emarginulata reticulata,

Eulima bilineata, .

distorta,

* polita,

Fissurella reticulata,

Fusus antiquus,

Islandicus, .

Kellia rubra, .

suborbicularis,

*Lacuna pallidula, .

* vincta,

Lamellaria perspicua,
*Leda caudata,
*Lepton convexum,

Lima hians, .

loscombii,

subauriculata,

Littorina littoralis,

littorea, .

- neritoides,

Dead shells, not uncommon.

( Good specimens from gravel
< bottom, at about 15 fathoms,
( north end of Holy Island.

( Concealed in the plant Lachina
< pigmcea, near the shore, Holy
( Island.

( In different parts of Lamlash
\ Bay.

( At the north end of Holy
Island, in about 10 or 15
fathoms. The nests of Lima
hians, with the live shells, are
very abundant in this lo-
cality; and they are worth
examining for Diatomacece.

278

MARINE ZOOLOGY OF ARRAN.

Littorina rudis,
*Lucina flexuosa,

* -- spinifera, .
*Lucinopsis undata,
*Lmtraria elliptica, .

Lyonsia Norvegica,
Mactra elliptica,

- solida,

- stultorum, .

- subtruncata,
Mangelia costata, .

- Leufroyi,

* - — linearis, .

* - nebula, .

* - septangularis,
-- teres,

* - turricula,
Modiola modiolus, .

* -- phaseolina,

*Montacuta bidentata,

- substriata,
*Mya areuaria,

* - truncata, »

Mytilus edulis,

Nassa incrassata, ..
- reticulata, .
*Natica alderi,
-- monilifera, .
- Montagui, .
*Nucula nitida,
-- nuaeus,

- radiata,
Odostomia conoidea,

-- cylindrica,

- decussata,

- excavata,

- interstincta,
-- plicata, .
-- rissoides,

- spiralis,
-- unidentata,

In the sands, low water,
Lamlash.

In deep water between Clach-
land Point and Holy Island.

M. rufa was dredged between
Fullarton'sRock and King's
Cross Point.

( Attached to the spines of
( SpaAangus purpureus.

Immature specimens every-
where.

MARINE ZOOLOGY OF AKRAN.

279

Ostrea edulis,

*0vulum acuminatum,
Patella athletica, .

pellucida, .

vulgata,

Pecten maximus, .
operculavis,

* pusio,

similis,

striatus,

An oyster-bed, Lamlash Bay,
Holy Island.

Holy Island.

»Lamlash Bay.

tigrinus,

— varius,

In the sandstone, low water,

Pectunculus glycimeris, .
Philine aperta,
— catena,

* -- scabra,

Pholas cnspata, . .

Pileopsis Hungarians,
Pilidium fulvum, .

* Pinna pectinata,* .

Pleurobranchus mem- J Under stones, low water, Holy

branaceus (?), . . ( Island.
*Pleurotoma gracilis,

- septangularis,
Psammobia Ferroensis, .
Puncturella Noachina, .
Purpura lapillus,
*Rissoa Beanii,

* - calathus,

* — - cingillus,

- costata,

* -- costulata,

* — - crenulata, .

* --- fulgida,

* -- inconspicua,

* -- interrupta, .

* -- labiosa,

* — — parva, . „

* - punctura, . *•

* Pinna pectinata was also dredged by the late Major Martin on
Skelmorlie Bank, Dear Wemyss Bay, Ayrshire coast.

280 MARINE ZOOLOGY OF ARRAN.

*Rissoa rubra,

* rutilabris, .

semistriata, .

— stnata,

- striatula,

— ulvae, .

- vitrea,
Zetlandica,

*Saxicava arctica,

* rugosa,

*Scalaria com munis,

Scaphander lignarius,
*Skenea divisa,

* nitidissima,

* planorbis, .

* rota, .

*Sphcenia Binghami,
*Syndosmya prismatica,

Tapes aurea, .

decussata,

virginea,

*Tellina crassa,

- donacina,

* tabula,

* incarnata. .

* solidula,

— tenuis,

m i , i f Deep water, between Hamil-

Terebratulacaput-serpentis, | JJ Rock and Holy Island.

*Teredo megotara, .

XT (In the wreck of the pier, Holy
Norvegica, . . Isknd

Thracia phaseolina,

• pubescens, .

Trichotropis borealis.
Trochus cinemrius,

— magus,

• millegranus,

Montagui,

pusillus, .

tumidus, .

umbilicatus,

undulatus,

ziziphinus,

Lam lash Bay. T. umbilicatus
is common along the shore
of Holy Island, also T.

MARINE ZOOLOGY OF ARRAN.

281

*Trophon Barvicensis,
* clathratus,

Turritella communis,
*Turtonia minuta, .

Venus casina,

fasciata,

gallina,

ovata,

striatula,

[ North end of Holy Island, in
from 10 to 20 fathoms.

TABLE II. — NUDIBRAXCHIATE MOLLUSCA.

Doris bilamellata, .

planata,

tuberculata, .

*Eolis alba,
* coronata,

Drummondi, .

* Landsburgii, .

Goniodoris nodosa, .

Lomanotus

flavidus,

Polycera quadrilineata,
Triopa claviger, .

Under stones at low water, in
many places, and not un-
common. D. planata, was
found in Lamlash Bay by
Mr. Alder.

( Under stones at low water, in
< many places, and not un-
I common.

Probably a new species,
dredged in Brodick Bay by
Dr. Greville and Dr. Miles,
in about 15 fathoms, be-
tween Invercloyand Gome-
gills ; but the specimen was
unfortunately lost before it
could be sent to Mr. Alder
for examination. It was
two inches in length, white,
with orange processes, and
of rare beauty. A sketch
taken by Dr. Greville was
forwarded to Mr. Alder.
L. flavidus was dredged in
1846 by Mr. Alder in Lam-
lash Bay.

282

MARINE ZOOLOGY OF ARRAN.

TABLE III. — CRUSTACEA.

Balanus balanoides,
Cancer Pagurus,

Carcinus Moenas,
*Ebalia Bryerii,

* Cranchii,

Pennantii, .

J The Cirripedia have scarcely
| been examined.
( The edible crab is common
around Arran.

Eurynome aspera, .

Galathea squamifera (?),
Gonoplax angulata,
Hippolyte varians, .

Homarus vulgaris, .

Hyas araneus,

* coarctatus, ... »

Inachus Dorsettensis,

*Lepas anatifera, .. .

Lithodes Maia, . - .
Pagurus Bernhardus,

Prideauxii, .

Palsemon squilla, .
Palinurus vulgaris,

Pandalus annulicornis, .
*Pinnotheres pisum,
Porcellana longicornis, .

platycheles, .

Stenorhynchus phalangium,

f Dredged by Dr. Greville in
\ Lamlash Bay.

Not uncommon in deep water.
( Several specimens obtained
< from deep water, Lamlash
( Bay.

J On the Ayrshire coast, abreast
^ of Arran.

( The common lobster is found
J among the rocks on different
j parts of the Arran coast
^ and at Holy Island.

J Found in Lamlash Bay on
\ floating wreck.
f In deep water, mid-channel,
\ off Arran.

Abundant everywhere.
J Abundant in Lamlash Bay,
|^ with Ada/nisia palliata.

Common in all the rock-pools.
f Obtained by the late Major
\ Martin from off Campbelton.

Among th e roots of Laminaria.
( Abundant under stones at low
water, Holy Island and
( Lamlash.

Dredged in deep water.

MARINE ZOOLOGY OF ARRAN.

283

TABLE IV. — ECHINODERMATA.

Amphidotus cordatus,

* roseus,

Asterias aurantiaca,

Asterina gibbosa,

Chirodota digitata, .
inhserens,

Comatula rosacea, .

Milleri, .

Cribella oculata,

rosea,

Echynocyamus pusillus, .
Echinus miliaris,

• sphcera,

Goniaster Templetoni,
Luidia fragilissima,
Ophiocoma bellis,

granulata,

rosula, .

Ophiura texturata, .

Palmipes membranaceus,

Sipunculus (?),

Spatangus purpureus,
Solaster papposa,

. endica,

Uraster glacialis,
• rubens,

The following species were all

obtained in Lamlash Bay.
Near Fullarton's Rock.

{In pools, north end of Holy
Island, &c.

J Dredged near the pier, Holy
\ Island.
Mull.

Abundant in Lamlash Bay,
near Holy Island, in from 8
to 15 fathoms. Solitary in-
dividuals have been traced
up the Clyde as high as
Hunter's Quay, Dunoon.
In Lamlash Bay, rare.

( This beautiful star-fish (two
I specimens) was dredged in
j deep water, north end of
[_ Holy Island.

Not uncommon in the bay.
Not uncommon in the bay.

f Good specimens, deep water,
1 near Hamilton Rock.

J One specimen found by Dr.
\ Landsborough.

Actinia bellis,

TABLE V. — ZOOPHYTA.

Jin the rock -pools, not un-

284

MARINE ZOOLOGY OF ARRAN.

Actinia coriacea,
crassicornis,

mesenibryanthemum,

( Under ledges
( pools, <kc.

of rocks, in

Adarasia palliata, .

Anguenaria truncata,

Anthea cereus,
Antennularia antenniua, .

— ramosa,
Cellepora pumicosa,

• ramulosa,

Skenei, .

Campanularia dumosa, .
Flustra foliacea, . .
Halecium halecinum,
*Halichondria panicea

suberea,

Laomedea geniculata,
Lepralia annulata, .

hyalina,

Malusii, .

Peachii,

trispiuosa, .

violacea,var. cru-

enta,

Lucernaria cyathiformis,

Flumularia pinnata,
Salicornaria farciminoides,
Sertularia tamarisca,

( Abundant in Lamlash Bay, at-
-/ tached to shells occupied by
{ Pagurus Prideauxii.
( First found by Dr. Lands-
\ borough.

On stones and dead shells, in
deep water, Lamlash Bay.

( Found by Dr. Landsborough
( same time as by Sars.

To these lists we have only to add two species, kindly
supplied by Mr. David Robertson —

CONCHIFERA. ECHINODERMATA.

Nesera cuspidata. Synapta inhserens.

85

ENTOMOLOGY OF AKBAN.

LEPIDOPTERA.

121. AMONG the Lepidoptera, or tribe of butterflies and
moths, occurring in the Isle of Arran, are several of rather
peculiar interest ; and a work treating of the Fauna of Arran
would be incomplete without some notice of them.

COLIAS EDUSA (the clouded yellow butterfly) is a species
which is very seldom seen so far north ; but its occurrence
near Lamlash was chronicled in the Zoologist for 1848, p.
1985.

EREBIA BLANDINA (the Scotch Argus) is a mountain
species of butterfly, which, thoiigh common in many northern
localities, is esteemed a prize by all collectors of the plains,
when first they meet with it. It is common on many of the
hill- sides in Arran.

EREBIA LIGEA is a species closely allied to BLANDINA,
which has been reported to occur in the Isle of Arran ; and
Mr. Curtis, the distinguished author of British Entomology,
assured me that the late Sir Patrick Walker told him he
had himself taken it in the island, in the second half of
August. No recent captures of this insect in Britain are
known; and the fact of its having ever been caught here
being much doubted, the species has ceased to figure in our
lists of indigenous species.

CCENONYMPHA DAVUS is common on boggy places at some
altitude above the sea.

Of the handsome genus of FRITILLARIES three species —

286 ENTOMOLOGY OF ARRAN.

ARGYNNIS AGLAIA, ADIPPE, and SELENE — are rarely met
iu the little glens, up the hill-sides; and in early spring
THECLA RUBI (the green hair streak) is frequent amongst
bramble bushes.

POLYOMMATUS ARTAXERXES is a species very likely to
occur in Arran, though I am not aware that it ever has
been found there.

ANTHROCERA NUBIGENA. — The capture near Oban of this
species of Sphhix, found also in Ireland, renders it certainly
not improbable that it may occur in Arran. Like many of
its congeners, it is excessively local, and may occur in some
limited spot, only on one hill-side, yet there be in profusion.
Among the family of the day-flying moths, the BOMBYCINA,
none of those which have been observed in the island are of
sufficient importance to deserve special notice.

Amongst the NOCTUINA, the pretty THYATIRA BATIS
(peach-blossom) is common ; CEI^ENA HAWORTHII is tolerably
plentiful ; CRYMODES EXULTS has once been taken on the face
of stone rocks ; and PLUSIA INTERROGATIONS is rather com-
mon than otherwise.

Amongst the GEOMETRINA which have been noticed here,
may be mentioned M.ESIA BELGIARIA, which flies in heathy
places ; OPORABIA FILIGRAMMARIA, LARENTIA C.ESIATA,
SALICARIA, and the mountain species of the pretty little
genus EMMELESIA, viz.: T.SNIATA, ERICETATA, and BLANDIATA,
and the pretty insect CARSIA IMBUTATA, of which the Scotch
specimens are so much more delicately marked than those
from the neighbourhood of Manchester.

Phibalapteryx lapidata. Hepialus hecta.

Cidaria russata — a peculiar Crambus latistrius.
and pretty variety. Habrostola triplasia.

Of the smaller species of Lepidoptera none of peculiar
interest have yet been noticed in Arran; but it must be
borne in mind that this portion of its natural treasures has
never been thoroughly explored.

ENTOMOLOGY OF ARRAN.

237

LIST OF LEPIDOPTERA COLLECTED IN 1836 BY DR. CONNELL:* —

Cynthia cardui.
Hipparchia blandina.

- poly dam a,

- pamphihis.
hyperanthus.

semele.

Polyommatus alsus.

Alexis.

Vanessa urticae.
Pontia brassicse.

rapse.
•napi.

Melitaea Euphrosyne.
Argynnis Aglaia.
Lycaana phlseas.
Arctia caja.
Cerura vinula.
Macroglossa stellatarum.
Minoa cbcerophyllata.
Anthrocera filipendulse.
Spilosoma menthastri.
Hypena proboscidalis.
Mamestra brassicae.

Hepialus velleda.

humuli.

Paimia crataegata.
Ophiusa lusoria.
Leucania pallens.
Plusia chrysitis.

gamma.

Episema cceruleocephala.
Leucania impura.
Lai'entia clienopodiata.
Fidonia atomaria.
Cabera exanthemata.

pusana.

Actebia porphyrea.
Xylina putris.
Harpalyce fulvata.

sylvatica.

ocellata.

Margaritia verticalis.
Botys forficalis.
Anarta myrtilli.
Pterophorus punctidactylus.
Nemeophila plantaginis.

* This list contains all that has been published up till this time
regarding the Entomology of Arran. It was furnished to Dr. Lands-
borough by the late Dr. Connell of the High School of Glasgow; and
was published first in the New Statistical Account of Scotland,
vol. v., and afterwards in Dr. Landsborough's Excursions to Arran.

288

THE BIRDS OF AKRAN.

122. The island of Arran, constituting the larger portion
of the shire of Bute,* contains many elements of a highly
attractive nature to some of the more important birds that
frequent the western counties of Scotland. From its diversi-
fied scenery, embracing picturesque glens and lofty mountains,
natural caves and moorland wastes — it gives ample shelter
to many species that are daily becoming scarcer in places
where the undue preservation of game has so much altered
the bird-life of our northern lands. With the exception of
the island of Skye, there is perhaps no part of the Western
Hebrides more remarkable for the sublimity of its bird-haunts
than Arran. The towering peaks of Goatfell and Ben Ghnuis
— homes of the eagle and ptarmigan — can nowhere be ex-
ceeded for grandeur and magnificence; yet, wild and savage
as they appear, they are not more impressive than the great
and silent glens with which they are intersected. Of these,
Glen Sannox, Glen Rosa, Glen Shirag, and Glen Catacol,
being the chief, are at once distinguished for beauty and sub-
limity. Some of the naked precipices enclosed in their
wildest recesses are inhabited by ravens and peregrines; and
in those places where the valleys expand into a romantic
softness, and are enriched by the copsewood which clothes
their gentle slopes, the smaller insessorial birds enliven every
hollow, and fill the air on summer evenings with their
music. Again, —

" Over the brow of the ferny hill,
Over the moorland, purple dyed,"

• The county of Bute consists of seven islands, viz., — Bute, Arran,
Great Cumbrae, Little Cumbrae, Inchmarnock, Holy Island, and
Pladda,

BIRDS OF ARRAN. 289

various birds of prey yet cast the shadow of their wings —
the merlin, that swift and courageous, though diminutive
plunderer, that strikes fear into so many birds larger than
itself ; the hen harrier, quartering the heath with keen and
regular scrutiny; the buzzard, perched on some gray boulder,
on the watch for helpless quarry; and the short-eared owl,
coursing the ground in open day, being all suggestive of the
inland heather wastes. Here, too, the ground offers a favour-
able summer encampment for the golden plover, lapwing,
redshank, and dunlin; while the various mountain tarns,
such as Loch Ghnuis, Loch Tanna, Dubh Loch, and Loch
Cnoc, encourage numbers of ducks and other waterfowl,
besides a few sea-birds — the greater and lesser black-backed
and common gulls being the chief inhabitants of these lonely
nurseries. Contrasted with such solitary abodes among the
dark and sombre mountains, the sandy shores and bays of
Arran form a pleasant resort; and on quiet days and calm
nights the shell-covered strands are sometimes crowded with
wading birds, whose noisy demeanour, as the tide recedes,
often causes the listener to wonder whence the babbling
sounds proceed. Sandpipers, redshanks, curlews, and gulls
all join in the general clamour. In some of the less fre-
quented sea lochs and bays, these piping hosts may be heard
even at midnight ; and when the sounds reach the ear from
a distance, the effect is not unpleasing. The caves, on the
other hand", which are numerous on both sides of the island,
are silent and gloomy — the only voices from within being
the cooing of rock doves, almost inaudible, however, through
the sullen boom of the waves. A few cormorants are found
here, and in one or two instances the gentle martin builds
its nest on the cliffs close at hand. At the south end of the
island the shelldrake breeds, — young and old, when gathered
together towards the close of summer, forming a conspicuous
congregation on the waters; and on hot and bright days,
when the sea is often like burnished gold, one of the best
bird sights on the Firth of Clyde is the appearance of small

u

290 BIRDS OF ARRAX.

groups of the great northern and black-throated divers off
shore in their brilliant summer dress, and calling to one
another in hoarse accents, as if giving the word to prepare
for the Icelandic or Hebridean lakes. From its nearness to
Ailsa Craig, Arran is visited by all the birds which are to
be found there — large flocks of guillemots, razorbills, puffins,
and other familiar tenants of that famous bird-hive being
seen almost constantly in Kilbrannan Sound, as well as along
the eastern shores of the island. It will be seen, therefore,
that the natural features of Arran are sufficiently diversified
to attract a very large proportion of the birds to be found in
Scotland, and that on this account the bird-student who may
make this beautiful island a temporary post of observation,
is enabled to acquire, within a comparatively small compass,
a knowledge of the habits of species ranging from the monarch
of its own Alps — the sea eagle — to the little storm petrel
whose shadowy form at twilight is occasionally detected by
the passing mariner.

123. Before giving a catalogue of species, it may not be
out of place to notice here what has hitherto been recorded
of the ornithology of Arran. The earliest notice which I
have been able to trace is extremely brief, and appeared in a
work on the Western Islands of Scotland, written about the
year 1703, by M. Martin, who states that " This isle affords
the common sea and land fowls that are to be had in the
"Western Isles." With this somewhat unsatisfactory account
naturalists remained content, until Pennant visited the island
about seventy years later, namely, in 1772. "The birds,''
says this writer,* " are eagles, hooded crows, wild pigeons,
stares, black game, grouse, ptarmigans, daws, green plovers,
and curlews." To these he adds the partridge, with the
remark that its presence there was a proof of the advance-
ment of agriculture. Pennant, it may be observed, speaks
of having seen eagles soaring over the naked summits of
Grianan Athol. Following these records, the Old Statistical
• Voyage to the Hebrides, Chester, 1774.

BIRDS OF ARRAN. 291

Account of the Parish of Kilmory (1793, vol. ix. p. 170)
contains a slight reference to one or two birds, by the
Rev. John Hamilton, who speaks of the wild swan being
found there in winter, the fieldfare in time of snow, and the
lapwing as a visitant for the months of May and June only.
In 1807, the Eev. James Headrick published a work on the
Mineralogy, Agriculture, and Antiquities of Arran, in
which, under the head of " Game," he has the following re-
marks : — " Among the feathered tribes, the capercailzie for-
merly abounded, but seems now to be extirpated. The black-
cock and grouse swarm in the most unbounded profusion; and
I suspect their further multiplication is prevented by not
having a sufficient proportion annually destroyed. Among
these a few quails occur; on the granite mountains the ptarmi-
gan has chosen his residence. Among the wild birds are
plovers of various species, starlings, and others of smaller
size, which seem to be migratory. A few partridges are
found at South End; and this bird can only propagate as
cultivation increases. The birds of prey are — eagles, kites,
hawks of various species, and carrion crows. The premiums
given by the late Duke for the preservation of game have
nearly effected their extirpation." In Mr. Headrick's book,
fi*equent mention is made of localities bearing Gaelic names
derived from their association with birds, as Castle Abhael
(ptarmigan's castle), Glen Chalmadale (glen of pigeons),
Torneadaneoin (heap of birds' nests), &c., shewing that in
early times the bird-life of the island had forced itself upon
the attention of the inhabitants.

The next account of the ornithology of Arran, written in
1837, and revised in 1840, includes a somewhat extended list
of species, and was ptiblished in the New Statistical Account
of Scotland, by the Rev. Allan M'Naughton, D.D., minister
of the parish of Kilbride. This writer also speaks of the
great abundance of grouse and blackgame — the latter
descending to the low grounds, and seriously annoying the
farmer in spring-time and harvest. Pheasants had been.

292 BIRDS OF ARRAN.

introduced a few years before, and were then spreading over
the whole of the parish. Eagles, falcons, hawks, hooded
and carrion crows, ravens, owls, and magpies, once numerous,
wei-e being rapidly reduced, through premiums given by the
Duke of Hamilton for their destruction ; and, as a conse-
quence, small birds had greatly increased, to the prejudice of
garden fruits. Seven years afterwards, viz., in 1847, the
Rev. D. Landsborough published a very interesting volume
on Arran and its Natural History, in which various allusions
are made to birds, though no list is given ; and in 1852
there appeared a second series of excursions by the same
author, containing a review of the birds of Ailsa, and casual
references to those of Arran and the Cumbrae Islands.
Since that time, with the exception of a short paper in the
Zoologist for 1866, by my friend Mr. Alston, nothing, so far
as I can trace, has been published relating to the subject.

In drawing up the following list, I have to express my
obligations to Mr. Halliday of Brodick, and Mr. Mackenzie
of Dougrie Lodge, from both of whom I have received much
useful information.

CATALOGUE OF SPECIES.
124. RAPTORES.

GOLDEN EAGLE (Aquila chrysaetos). All the older writers
on the zoology of Arran agree in referring to more than one
species of eagle as nesting in the island. It is not easy, how-
ever, to say with certainty at what time the golden eagle
ceased to rank as a native, and became, what it now in, a
rare and uncertain visitant. The locality is quite as attrac-
tive as it was a hundred years ago; and, so far as I can learn,
no eagles have been shot or trapped for a very long period.
The bird, therefore, must have been scared at a time when
premiums were persistently offered for its destruction — a

BIRDS OF ARRAN. 293

persecution which was doubtless the cause of the desertion of
the eyries which formerly existed on Ailsa Craig and one or
two localities on the opposite mainland. Two eyries of this
eagle in Argyleshire were robbed in April 1872. Each con-
tained two eggs ; one pair, exhibited unblown at a meeting of
the Natural History Society of Glasgow shortly afterwards,
were large, heavy, and handsomely spotted ; the other two,
which were submitted to me before being blown, were equally
fine, though not so large. Both nests were taken in districts
within sight of Goatfell.

WHITE- TAILED EAGLE (Ilaliaetus albicilla}. Though this
bird has suffered the same treatment as the golden eagle, it
appears to be better able to withstand the pains and penal-
ties of keepers throughout the range of its Hebridean haunts.
As a rule, its eyrie is more difficult of access than that of its
ally, and the bird itself certainly commits less mischief on
sheep farms and game preserves. In Arran it continued to
breed regularly in Catacol Glen until 1849 — in which year,
as Mr. Alston informs me, Robert Mackay, keeper, took a
young bird and reared it. Sometime, during the ensuing
winter the ledge of rock on which the eyrie was placed gave
way, and the eagles — probably viewing the mishap as a
timeous warning — left the district. In the spring of 1870 a
pair of these birds built a nest in the same glen, and the
female sat for some time on two eggs, but afterwards deserted
them. Mr. Mackenzie, one of the Duke of Hamilton's
keepers, residing at Dougrie Lodge, on the west coast of
Arran, informs me that he saw two white-tailed eagles there
in November 1871. I am happy to be able to add that
strict orders have been given that eagles are not to be
destroyed in any part of the island.

OSPREY (Pandion haliaetus). Mr. Alston has written to
me as follows regarding the occurrence of the osprey in
Arran : — " This fine bird formerly bred in North Glen
Sannox. According to Mr. Halliday, the eyrie was robbed
by Dr. Martin Barry in 1846, and again in 1847. Since that

294 BIRDS OF ARRAN.

date it has not bred in Arran." I may here remark that the
eggs said to have been taken by Dr. Barry were sold iu
London by Mr. J. C. Stevens, in April 1861. A copy of the
catalogue from which I make the following extract is now
before me : — " The osprey now no longer breeds in our
Highland lochs, and the bird itself is becoming extremely
scarce. One nest was taken in the presence of the late pro-
prietor in the Isle of Arran from the side of a precipice
overhanging North Sannox Glen (twenty fathoms over), only
six fathoms less than Goatfell in height." As I do not mean
to refer to this catalogue again, I have only to add that the
foregoing extract is sufficient to shew that it must have been
drawn up by some one who had never been in Arran, and
who certainly must have been thoroughly ignorant of the
habits of British osprey s.

PEREGRINE FALCON (Falco peregrinus). Breeds every year
in various localities. The principal stations are Holy Island
(where the nest is nearly inaccessible), Maoldon, and Glen
Dubh. Until within recent years there was a nest on a cliff
in the Lesser Cumbrae; but through repeated disturbance
this station has been abandoned The following extract,
from a Historical Memoir of the Family of Eglinton and
Winton, by John Fullarton, Esq., published in 1864,
shews that the peregrines frequenting this haunt, and that
of the opposite cliffs in Ayrshire, had been reckoned of
some value: — "In 1609 a complaint was brought before
the Privy Council by the Captain of Dumbarton Castle,
that Robert Hunter of Hunterston, and Thomas Boyd,
provost of Irvine, had gone to the Isle of Comra,
with convocation of the lieges, and tane away all the
hawks therein. It was decerned, 'That all the hawks
quhilk bred on the said isle do properly belong to the king,
and ocht to be forthcomand to his majestic, . . . and dis-
charges the said Robert Hunter and all others from meddling
therewith.' That this alludes to the island of the Lesser
Cumbrae, there need be no doubt, the greater island being

BIRDS OF ARRAN. 295

quite unsuitable as a habitude for these birds ; and Hunter,
it is not improbable, may still have conceived some right or
use for what he did. The hawks to some extent still main-
tain their footing here, as likewise on the opposite majestic
precipice of Ardueilbank. Preparatory to the visit of George
IV. to Edinburgh, in 1822, the redoubted old falconer,
Fleming of Barochan, in the barony of Renfrew, sent his
trusty and kindly old trainer, John Anderson, to procure
eyesses to be trained and presented on that notable occasion ;
and it was understood he was at least successful in obtaining
a number from the eyries of Ardneilbank."

I examined a very fine female peregrine which was shot
in the Island of Bute in April, 1872.

HOBBY (Fcdco subbuteo). Mr. Mackay, resident at Loch
Ranza, informs me that he has taken the nest of this bird in
Arran, in company with the late Dr. M. Bany. The bird has
been accurately described to me ; and I see no reason, for
rejecting the information. In 1857 a specimen of the
hobby, freshly set up, and said to have been procured in
Arran, was shewn to me for identification. Being a rarity, it
was, after passing through various hands, lost to the original
owner, and I have since been unable to trace its ultimate
fate.

MERLIN (Falco cesalori). This beautiful and active little
bird of prey is permanently resident in Bute and Arran,
breeding in both islands ; several nests are met with yearly
by Mr. Halliday and his assistants in the coxirse of their
walks over the moors.

KESTREL (Falco tinnunculiis). A very common species,
breeding in suitable localities throughout all the islands. In
Cumbrae the nests are chiefly found along the range of red
cliffs to the north of Millport ; but these haunts are likely to
become deserted on account of the extension of house-building
in that direction.

SPARROW-HAWK (A emptier nisus). Common. Mr. Alston
informs me that a gentleman of his acquaintance told him of

296 BIRDS OF ARRAN.

having a few years ago found a sparrow-hawk's nest near
Loch Ranza, placed on the ground.

KITE (Milvus vulgaris). Is now entirely extinct on all
the islands. It is referred to by Headrick in his work,
already spoken of, and also by the writer of the Statistical
Account of the Pariah of Kilmory, published in August
1840. In a manuscript account of the Natural History of
Bute, written by John Blain, commissary of the isles about
the year 1800, it is likewise specially mentioned and alluded
to as being dark in colour.*

* This manuscript, entitled a History of the Island of Bute, and
forming a bulky volume of nearly 500 pages, formerly belonged to Mr.
J. Eaton Reid, of Glasgow, and was presented by that gentleman to
the Burgh of Rothesay. The following list of birds is given by the
author ; and I have the pleasure of inserting it here through the kind
permission of Mr. Wilson, Town Clerk : — " Of fowls and birds abiding
or coming in their seasons, or waterfowl on or visiting the coast, we
have the gallina or hen, the turkey, tame goose, tame duck, red
game, blackcock and heath hen, partridge, blackbird, throstle,
green and gray linnets, goldfinch, bullfinch, starling, lark, corn-
bunting, house and hedge sparrows, sitta, skipper, oxeye, yellow-
hammer, wagtail, redbreast, blackcap, mosscheeper, wren, fieldfare,
cuckoo ; domestic, rock, and wood pigeons, land and water rails,
curlew, woodcock, large and small snipe, gray and green plovers,
raven, crow, rook, jackdaw, little water crow with a white breast,
magpie, jaepy, kite of a dark-brown colour, others light-gray or nearly
white, hawks of many kinds, eagle, falcon, the common owl, common
house swallow, the Held swallow, wild goose, solan goose, barnacle,
common wild duck, widgeon, teal, black and white diving fowl,
called here Danes, diving fowl with black and white breast, herons,
which breed in considerable numbers among the planting at Mount
Stuart ; allan hawk, with white breast and speckled back and neck ;
Heart, of dark -brown colour; scale drakes and ducks near as large as a
goose ; another sort of them, about the size of a common duck, which
the people have sometimes tamed by procuring their eggs and hatch-
ing them under a tame duck ; sundry varieties of the larus or until
kind ; sandlarks. redshanks, marrats, Ailsa black seacocks, sea-
parrots ; water hens, having scalloped webs along each toe. These
hens are all black or very dark -brown, only a white spot at the root
of the bill, but they are not to be frequently met with." It may be

BIRDS OF ARRAN. 297

COMMON BUZZARD (Buteo vulgaris). Was quite common in
Arran until 1851 or 1852, since which time it has become
scarce. One was trapped at Brodick in 1865, and other two
at the same place in 1871.

Obs. — I have not been able satisfactorily to trace the
rough-legged buzzard, honey buzzard, or marsh harrier, in
any of the islands, though I have no doubt they were occa-
sionally met with before the increased protection to game led
to a general destruction of raptorial birds.

HEN HARRIER (Circus cyaneus). Breeds regularly. Mr.
Halliday and Mr. Mackenzie find several nests every year.

LONG-EARED OWL (Otus vulgaris]. A few pairs only in the
plantations of younger growth.

SHORT-EARED OWL (Otus brachyotos}. Breeds regularly,
and is seen almost daily hunting the moors for prey in the
heat of noon. Mr. Halliday, who has frequently found the
nest, took one with four eggs in April, 1872.

WHITE OR BARN OWL (Strix flammea). Sparingly met
with in different localities throughout the islands of the
county of Bute.

TAWNY OWL (Syrnium stridula). Not uncommon in all
the wooded islands.

Obs. — It is quite possible that the Snowy Owl and
Hawk Owl may both find their way at times to some of the
islands, as these birds have repeatedly been met with in the
Firth of Clyde. Such examples, no doubt, come to this
country by means of ships, and make for the shore at the
first opportunity. A specimen of the hawk owl was taken
in the flesh to a bird-stufFer in Greenock in December, 1871.

125. INSESSORES.
SPOTTED FLYCATCHER (Muscicapa grisola). Very common.

added that the allan hawk is the great northern diver, which in some
parts of Ayrshire is still known by that name; and that the marrat is
the razorbill. I am unable to find out what the writer has meant by
the divers called Dane*.

298 BIRDS OF ARRAN.

COMMON DIPPER (Cinclus aquaticus). Common on all the
streams.

MISSEL THRUSH (Turdus viscivorus). Very abundant
within the last twenty years.

FIELDFARE ( Turdus pilaris). Plentiful as a winter visitant.

SONG THRUSH (Turdus musicus). Abundant, and a most
useful and interesting songster.

REDWING (Turdus iliacus). Common, arriving with the
fieldfare.

BLACKBIRD (Turdus merula). Has increased greatly
during the last thirty years, and is now extremely abundant.

RING OUZEL (Turdus torquatus}. Very numerous in the
moors, where it breeds, descending to the low grounds in
autumn. In the Rev. Dr. M'Naughton's catalogue, this
bird brings up the rear of fruit-destroying thieves as " the
most numerous and pestilent of the whole" — & character
which it had evidently acquired in the manse garden.

GOLDEN ORIOLE (Oriolus galbula). A specimen of this
beautiful bird was shot at Loch Ranza in 1807, and is
referred to by the late Dr. Fleming in his History of British
Animals.

HEDGE ACCENTOR (Accentor modularis). Very common.

REDBREAST (Erytfiaca rubecula). Abundant.

REDSTART (Phcenicura ruticilla). A few pairs in suitable
localities.

STONECHAT (Saxicola rubicola). Abundant — a few remain-
ing in winter.

WHINCHAT (Saxicola rubetra). Very common.

WHEATEAR (Saxicola cenanthe). Very abundant.

SEDGE WARBLER (Salicaria phragmitis). Abundant, and
appears to be yearly spreading through the island in greater
numbers.

BLACKCAP WARBLER (Curruca atricapilla). Sparingly
distributed in cultivated districts.

GARDEN WARBLER (Curruca hortensw). Scarce.

COMMON WHITTEHROAT (Curruca cinerea). Abundant.

BIRDS OF ARRAN. 299

WOOD WARBLER (Sylvia sylvicola). Sparingly found in
sheltered woods. I have also seen it in Bute and Cumbrae.

WILLOW WARBLER (Sylvia trochilus). In profusion in all
the woods, gardens, and roadside thickets.

Obs. — I have not yet succeeded in getting specimens of
the grasshopper warbler, the lesser whitethroat, or the chiff-
chaff in Arran, though I have no doubt they are all to be
found there. I have seen and shot the chiff-chaff in the
Cumbrae islands.

GOLDEN CRESTED WREN (Regulus cristatus). Found only
within the last forty years. It is mentioned by the writer
of the Statistical Account in 1837. It now occurs in great
numbers in many places throughout the county group, but is
migratory.

GREAT TIT (Parus major"). Yery common.

BLUE TIT (Panes cceruleus). Very common.

COLE TIT (Parus ater). Very common.

LONGTAILED TIT (Parus caudatus). Very common.

Obs. — Although the marsh tit may be reasonably looked
for in Arran, as well as the other islands, I have never yet
met with it.

PIED WAGTAIL (Motacilla yarrellii). Common, and per-
manently resident.

GRAY WAGTAIL (Motacilla boarula). Common, and per-
manently resident.

RAY'S WAGTAIL (Motacilla rayi). Common summer
visitant.

TREE PIPIT (Anthus arboreus). Common summer visitant

MEADOW PIPIT (Anthus pratensis). Abundant all the
year.

ROCK PIPIT (Anthus petrosus). Very numerous on all
the shores.

SKYLARK (Alauda arvensis). Abundant.

SNOW BUNTING (Plectrophanes nivalis). Large flocks seen
every winter.

COMMON BUNTING (Emberiza miliaria). Very common.

300 BIRDS OF ARRAN.

BLACKHEADED BUNTING (Emberiza schceniculus). Com-
mon.

YELLOW HAMMER (Emberiza citrineUa). Abundant

CHAFFINCH (Fringilla coelebs). Abundant.

BRAMBLING (Fringilla montifringilla). A few found in
winter in stackyards. Much less numerous here than in the
east of Scotland.

HOUSE SPARROW (Passer domesticus). Most abundant, and
yearly increasing.

GREENFINCH (Coccothraustes Moris). Yery common.

GOLDFINCH (Carduelis elegans). Not uncommon in the
various islands.

COMMON LINNET (Linota cannabina). Abundant, espe-
cially near the coast.

LESSER REDPOLE (Linota linaria). Rather a scarce species.

TWITE (Linota montium). Plentiful.

BULLFINCH (Pyrrhula vulgaris). Not yet common, but
increasing in suitable places.

COMMON CROSSBILL (LSxia curvirostra). Yery rare. In
1851, the late Dr. Lorrain, of Glasgow, was shewn a number
of female crossbills by a keeper in the Island of Bute, who
had shot them as curiosities, believing them to be greenfinches
with twisted beaks!

STARLING (Sturnus vulgaris). Very abundant. I have
found it breeding in great numbers at Lamlash, ill earth
embankments, in holes like those made by the sand martin.
Some years ago a pair built their nest in a hole of the pier
at Lamlash, almost within wash of the waves.

CHOUGH (Fregilus gracitlus). Now extinct in Arran.
Mr. M'Kenzie informs me that one of his men shot a pair in
1863, at Kilpa trick, on the west side of the island, but that
no choughs have been since seen in any part of Arran.

RAVEN (Corvtu corax). Still maintains its ground, not-
withstanding the means used for its destruction. It breeds
on Holy Island and in other places, four or five nests being
obtained yearly.

BIRDS OF ARRAN. 301

CARRION CROW (Corvus corone). HOODED CROW (Corvus
comix). Both these birds exist in some numbers, and inter-
breed here as at other places in the West of Scotland, with
like results.

ROOK (Corvus frugilegus). Common, and on the increase.

JACKDAW (Corvus monedida). Exceedingly abundant,
breeding in holes and crevices on the precipitous mountains,
where its nests are quite inaccessible. I have nowhere seen
this bird so numerous. It is likewise increasing rapidly in
the other islands.

MAGPIE (Pica caudaki). Now quite shot down iu Arran.
The last bird was killed at Dougrie about ten years ago, and
it was supposed to have come across from Cantire during a
westerly gale.

Obs. — The Jay is likewise extinct, none of the present
keepers having seen one during an experience in Arran of
nearly fifteen years. In Mr. Blain's MSS. it is mentioned
as being found in the Island of Bute.

COMMON CREEPER (Certhia familiaris). Common in suit-
able places.

WREN (Troglodytes vvdgaris). Abundant. Some years
ago, when walking from Brodick to Lamlash by the old road,
I found in the broken banks of a stream several wrens' nests,
cup-shaped like that of a swallow ; they were all placed on
tree-roots under the edge of the projecting turf, and in two
of them I found the bird sitting. One of these I caught
with my hand as it sat on the nest.

CUCKOO (Cuculus canorus). Abxindant even in the bleak-
est heaths of all the islands.

Obs. — Though the Arran streams are likely enough places
to attract the kingfisher, I am assured by Mr. Halliday and
Mr. Mackenzie that it is never seen in the island.

SWALLOW (Hirundo rustica). Very common. Mr. Alston
informs me that it builds in the face of the cliffs on the west
side of the island, where there are few houses.

MARTIN (Hirundo urbica). Also common, breeding in
colonies on the cliffs along the west coast.

302 BIRDS OF ARRAN.

SAND MARTIN (llirundo riparia). Common — arriving
much earlier than the other swallows.

SWIFT (Cypsdus apus). Very common.

NIGHTJAR (Caprimulgus Europceus). Very common,
breeding in many of the moorlands. Mr. Alston writes to
me, that having been told of the goshawk breeding annually
near Laggan, he had taken the trouble to apply to Mr.
Halliday for further information, and had found that the
nightjar goes by that name in Arran.

126. RASORES.

WOOD PIGEON (Columba palumbus). This bird is becom-
ing very numerous on all the wooded islands included in the
county of Bute. In Arran especially it is rapidly increas-
ing ; and I have noticed the same increase in Cumbrae during
the last twelve or fourteen years.

ROCK DOVE (Columba lima). Very numerous in the sea
caves. The breed, however, is much mixed in some places.
Mr. Duncan C. Brown informs me that he has frequently
seen these birds perched on leafless trees, in the neighbour-
hood of their haunts, in Arran.

PHEASANT (Phasianus colchicus). Was introduced into the
island of Arran about thirty-five years ago, and is now mul-
tiplying satisfactorily.

CAPERCAILZIE (Tetrao urogallus). In 1847, when Dr.
Landsborough wrote his work on Arran, capercailzies had for
some time been re-introduced into the island, but were then
kept in an enclosure made for their reception near Brodick
Castle. Since that time they have had their full liberty, and
are now tolerably numerous in that part of the island,
breeding freely in the open grounds, where the woods afford
them a ready shelter. Mr. Halliday informs me that he
often sees from thirty to forty capercailzies in the course of a
day's walk, and that he can always calculate on finding about
twenty nests in a season.

BLACK GROUSE (Tetrao tetrix). Most abundant It has

BIRDS OF ARRAN. 303

long been the subject of special protection. Writing in 1703,
Martin says, " The black cock is not allowed to be killed
here without a licence; the trangressors are liable to a fine."

RED GROUSE (Lagopus Scoticus). Is found in very great
numbers in Arrau especially, and is there remarkably tame
during the close season.

PTARMIGAN (Lagopus vulgaris). Has been long known as
a native of Arran. All the older writers mention it as an
inhabitant of the granite mountains. For many years it has
been on the decline, and attempts have recently been made
to restore the bi'eed by importing birds from Inverness-shire.
Mr. Halliday and Mr. Mackenzie agree in saying that there
are still a few on Ben Ghnuis and Goatfell.

PARTRIDGE (Perdix cinerea). Appears to have been intro-
duced into the island of Arran previous to Pennant's visit, in
1772. It is also mentioned as one of the birds of Bute by
Mr. Blain, whose unpublished account was written in 1800.
In both islands it is now fully established.

QUAIL (Coturnix vulgaris). Mentioned by Headrick,
in his work on Arran, in 1807. It is almost unknown at
the present day, though in Ayrshire, on the opposite main-
land, it is a regular summer visitant.

127. GRALLATORES.

GOLDEN PLOVER (Charadrius pluvialis). Very common,
breeding on all the hills, and frequenting tbjfc shores, at low
tides, during the autumn months.

Obs. — I have utterly failed to trace the dotterel (Ch.
morinellus) in any of the western islands.

RINGED PLOVER (Charadrius hiaticula). A common species
on all the islands, where it breeds.

SANDERLIXG ( Calidris arenaria). This species only appears
in the end of July or beginning of August, and remains no
longer than a week or ten days. I have failed to distinguish
it on the Arran shores in spring-time.

GRAY PLOVER (Squatarola cinerea). Occurs "ifc small

304 BIRDS OF ARRAN.

numbers in autumn only. I have never observed it in
spring.

LAPWING ( Vanellua cristatus). Formerly a summer visitant
for two months, after which it immediately disappeared. It
is now, however, seen from April to September.

TURNSTONE (Strepsilas interpres). This bird lingers every
year on the shores of Bute and Arran, and also on the shores
of the Kyles, until the end of May, by which time it has
nearly assumed the breeding plumage, and is then a most
attractive bird.

OYSTER-CATCHER (Hcematopus ostralegus). This beautiful
wader is common on all the islands, breeding in suitable
places in Bute, Arran, the Cumbraes, Pladda, and Inch-
marnoch. I have found the eggs on the Lesser Cumbrae,
near the ruins of the old castle.

COMMON HERON (Ardea cinerea). Common on all the
islands. There are two heronries within the county limits, —
one at Brodick, in Arrau, with a considerable number of
nests, the other at Mountstewart, near Rothesay, in Bute.
The last named has been long known, and is mentioned in
Mr. Blain's manuscript.

BITTERN (Botaurus stellaris). Very rare. A specimen
killed near Whitehouse, Arran, about the year 1835, is
referred to by the writer of the New Statistical A ccount of t/ie
Parish of Kilbride. I suspect it is now never seen.

CURLEW (Jfuifpnius arquatd). Abundant in all the islands.

WHIMBREL (Nvme.nius plueojws). Only a stray visitant
early in May, and not often obtained, its stay being of short
duration.

REDSHANK ( Totanus calidris). Breeds in limited numbers.
In autumn it is very common along all the shores.

COMMON SANDPIPER (Totanus hypoleucos). Very common
everywhere.

GREENSHANK (Totanus glottis). Only a stray winter
visitant.

BAR-T4ILED GODWIT (Lirnosa rufd). Occurs in small

BIRDS OF ARRAN. 305

numbers in company with gray plovers in the autumn months,
after which it disappears. I have never found the black-
tailed godwit in the district, nor the ruff, though it is possible
that both may have been overlooked.

WOODCOCK (Scolopax rusticola). A very plentiful bird in
Arran. It breeds in considerable numbers every year.

COMMON SNIPE (Scolopax gallinago). A very common
species in all the islands. I have observed great numbers in
Cumbrae.

JACK SNIPE (Scolopax gallinula). Common in winter, but
strictly migratory.

CURLEW SANDPIPER (Tringa subarquata). Rare. It
arrives in August, but remains only a very short time; and
unless carefully looked for, it is apt to be missed.

KNOT (Tringa canutus). Of late years this bird has be-
come more numerous in autumn. On its arrival it is found
on all the sandy shores, but only for twelve or fourteen days,
when it disappears. Great numbers of these birds were shot
near Ayr in August, 1871.

DUNLIN (Tringa variabilis). Plentiful, breeding on all the
hills of moderate elevation in the district.

PURPLE SANDPIPER (Tringa maritima). An autumn
visitant only, occurring a little later than the sanderling and
curlew sandpiper, but equally uncertain as to the length of
its stay. I have on one or two occasions found it in spring,
lingering till the middle of May, in the Kyles of Bute.

Obs. — Neither of the phalaropes can be included in this
list, though it is probable that both have been overlooked.

CORNCRAKE (Crex pratensis). Abundant.

WATER RAIL (Rallies aquaticus). Not plentiful, though
far from rare.

MOOR-HEN (Gallinula chloropus). Very common.

COOT (Fulica atra). A few pairs in various localities.
This bird is mentioned in Blain's MSS. as a species found in
Bute towards the end of last century.

306 BIRDS OF ARRAN.

128. NATATORES.

GRAY-LAG GOOSE (Anser ferus). Mr. Mackenzie informs
me that this species is met with occasionally on the lochs.
It is also mentioned in Blain's MSS.

Obs. — Though I have no doubt of the bean goose, the
pink-footed goose, white-fronted goose, the bernicle, and
brent goose, all occurring occasionally' within the limits of
the county of Bute, I have no authentic information of any
specimens having been killed and examined during the last
ten years. The older writers vaguely speak of wild geese
occurring in their parishes, without being able to distinguish
the species. It may be mentioned, however, that the
bernicle is catalogued by Mr. Blain, shewing that it at
least had been recognized.

HOOPER or WILD SWAN (Cygnus ferus). None of these
birds are ever now observed in Arran. Writing in 1793
from the parish of Kilmory, the Rev. J. Hamilton says :
" We have transient visits of the swan in winter." Wild
swans, however, regularly visit Renfrewshire and some
parts of Ayrshire, and they may occasionally fly past Arran,
as in days of yore. I examined six specimens of Bewick's
swan, from Ayrshire and Argyleshire, in January and
February, 1871, and other five were about the same time
procured in Dumfriesshire. Later still — namely, in October,
1871 — I saw two wild swans flying at a moderate elevation,
and apparently crossing from Arran towards the Heads of
Ayr.

COMMON SHELLDRAKE (Tadorna vulpanser). This beauti-
ful bird is plentiful at the south end of Arran, where it
breeds. It is also frequently seen off Pladda. In Mr. Blain's
account it is likewise included as having been found in Bute
at the time he wrote.

WILD DUCK (Anas boschas). Abundant throughout the
district.

TEAL (Anas crecca). Common; a few breed in the
marshes.

BIRDS OP ARRAN. 307

WIDGEON (Anas penelope). Common in winter.

EIDER DUCK (Somateria mollissima). Rare; occasionally
finds its way into Kilbrannan Sound.

VELVET SCOTER (Oidemia fusca). Very rarely seen.

COMMON SCOTER (Oidemia nigra). Yery rarely seen.

POCHARD or DUN BIRD (Fuligula ferina). Occasionally
met with in the lochs of Arran and Bute.

SCAUP DUCK (Fuligula martta). Of less frequent occur-
rence than the preceding species.

TUFTED DUCK (Fuligula cristata). Occasionally procured
in the lochs of Arran and Bute.

GOLDEN-EYE (Fuligula clangula). Not unfrequent in
immature plumage. Old birds may be said to be rare in
the two chief islands.

B/ED-BREASTED MERGANSER (Mergus serrator). I cannot
find any authentic instance of this bird having bred in either
Bute or Arran, though I have no doubt it is resident in the
last named island.

GOOSANDER (Mergus merganser). This handsome bird is
yearly becoming more plentiful in the Firth of Clyde.

RED-NECKED GREBE (Podiceps rubricollis). Uncommon.

SCLAVONIAN GREBE (Podiceps cornutus). I have seen
pairs of this bird in calm days off Arran in the beginning of
summer. Specimens in full summer plumage have been
sent this year (1872) from the outer Hebrides.

LITTLE GREBE (Podiceps minor). Common on all the
lochs, where it breeds.

GREAT NORTHERN DIVER (Colymbus glacialis). This
large and handsome diver is frequently seen off Arran at the
end of spring. Some birds linger till the first week of June,
and are then in their brilliant breediug plumage. A most
beautiful specimen was shot off Carradale, in Kilbrannan
Sound, in the last week of May, 1872.

BLACK-THROATED DIVER (Colymbus arcticus). The same
remarks apply to this as to the preceding species. I have
seen specimens caught as late as May 29th.

308

BIRDS OF ARRAN.

RED-THROATED DIVER (Colymhus septentrionalis). Of
frequent occurrence, though I have not heard of its nest
having been taken in any part of the district. Specimens in
full summer dress are occasionally seen off shore in May.

COMMON GUILLEMOT (Uria troile). Abundant in summer
all round the islands. The variety known as the ringed
guillemot has also occurred in Kilbrannan Sound — Mr. J.
Gilmour having shot two specimens there in April, 1868.

BLACK GUILLEMOT (Uria grylle). Tolerably common,
especially in Kilbrannan Sound. Eare off Bute and in the
Kyles. A few pairs nest on the west side of Arran, where
I have seen them in full breeding dress.

LITTLE AUK (Mergulus melanoleucos). Very rare. I
have seen but two or three specimens taken in the district,
and these after heavy gales. One was taken at Millport, in
Cumbrae, three or four years ago, and is now in the collection
of Mr. David Robertson.

PUFFIN (Fratercula arctica). Abundant all round the
coasts, appearing in February, and leaving in September.

RAZORBILL (Alca torda). Very numerous off shore, where
it sometimes congregates in hundreds at the close of the
breeding season.

GREAT AUK (Alca impennis). I may here be allowed to
refer to a history of this bird as a Scottish species in the
Birds of the West of Scotland, in which reference is made to
the fate of the specimen, procured by Dr. Fleming in the
outer Hebrides in 1821. This bird having escaped from its
owner off the Island of Pladda, Arran has the credit of being
associated with the last of the great auks seen alive in this
country. I have only to add, that all my efforts to trace the
history of the specimen found dead at Gourock have proved
fruitless.

CORMORANT (Phalacrocorax carbo). Very common in the
firth. There appears to be no breeding station for this bird
in any of the islands.

SHAG (P/ialacrocorax graculus). Also common. The

BIRDS OF ARRAN. 309

nearest breeding places are Ailsa Craig and the Mull of
Cantire.

GANNET or SOLAN GOOSE (Sula alba). Very abundant
throughout the summer months off all the islands. It occa-
sionally comes into the sheltered bays of Arran, Bute, and
C umbrae, and its appearance in such places generally, if not
always, forebodes a storm. Some years ago, during a terrific
gale, two of these birds dashed themselves against the lantern
of the lighthouse on Pladda, and caused the utmost conster-
nation among the keepers, one of whom declared the end of the
world was at hand, as he had seen angels with outspread wings
looking wistfully through the spray-covered glass. Daylight,
however, revealed the fact that they were but solan geese.

SANDWICH TERN (Sterna, Boysii}. I have seen this bird
passing Lamlash on one or two occasions. It is easily
distinguished from the commoner species while in flight.

ROSEATE TERN (Sterna Dougallii). First made known as
a British bird in 1812, by Dr. M'Dougall, who discovered a
colony of roseate terns on two small rocky islets, called the
Allans, in the Bay of Millport, Cumbrae. This colony existed
until 1848 or 1849, when, through repeated molestation, the
birds dwindled to a few pairs, and were afterwards completely
extirpated. In 1859 I found that another colony had
established itself on a similar group of islets off Carradale, in
Kilbrannan Sound, in company with a number of common
terns. A careful search within the limits of the county
would doubtless lead to the discovery of other settlements.

COMMON TERN (Sterna hirundo). Numerous throughout
the summer months.

ARCTIC TERN (Sterna arctica). Very common in the Firth
of Clyde, and often seen in the bays, fishing.

LESSER TERN (Sterna minuta). Occasionally met with.
I know of no breeding station in the district.

Obs. — Though I have not detected the black tern in Arran
or any of the other islands, I have no doubt it has been
passed over, especially in the plumage of the first year.

310 BIRDS OF ARRAN.

BLACKHEADED GULL (Larus ridibundus). Very common.
It has not, however, been recognized as breeding in any of
the islands.

KITTIWAKE (Larus tridactylus). Plentiful along the
coast, and remaining throughout the winter in considerable
numbers.

IVORY GULL (Larus ebumeus). A specimen of this rare
gull was shot near Lamlash in September, 1866.

COMMON GULL (Larus canus). Plentiful. It breeds on
Loch Ghnuis, and also on Loch Tanna.

ICELAND GULL (Larus Icelandicus). Mr. Thompson, in
his work on the Birds of Ireland, mentions that his friend
Mr. Sinclaire had seen six or eight Iceland gulls near
Brodick, in Arran, many years ago. They were extremely
tame. In the Birds of the West of Scotland I have stated
that this species is a regular visitant to the Clyde, and I
still entertain that belief. A specimen in the plumage of
the first year was shot on Gourock Moor, and sent to Mr.
Thomson, of the Kelvingrove Museum, Glasgow, on 20th
March, 1872. The keeper who shot it said there were others
in its company.

LESSER BLACK-BACKED GULL (Larus fuscus). Very com-
mon. It breeds on various lochs in Arran. About the last
week of June the old birds assemble in the sandy bays, while
the young ones remain on the moors to forage for themselves.
I observed twenty-four of these gulls on the beach at Lam-
lash on the 26th June. They were closely huddled together,
and I walked up to within twenty paces. They then uttered,
while sitting, a series of plaintive notes like gaw or gow, and
continued talking in that manner until I put them up.
Their cry then changed into a laughing chuckle as they
slowly dispersed.

HERRING GULL (Larus argentatus). Plentiful in all the
islands, especially in autumn, after it has quitted Ailsa Craig,
where it breeds in great numbers.

GREAT BLACK-BACKED GULL (Larus marinus). Tolerably

BIRDS OF ARRAN. 311

common. It breeds, though not in great numbers, in some
of the lochs of Arran. Mr. Mackenzie informs me that he
has caught the young ones among the heather as late as the
middle of August.

Obs. — I have frequently, in some of the sheltered bays
which are fringed with wood, seen flocks of gulls of
various species engaged in catching flies and other insects,
while at a considerable height in the air. Sometimes the
birds were near enough for me to see the gulls snapping
with open mandibles at the insects, like accomplished fly-
catchers. In September, 1867, when on the west side of
the island of North Uist, in the outer Hebrides, I saw
nearly a thousand sea gulls of all kinds hovering over the
heads of a large congregation of persons asembled for worship
in the open air. The weather was extremely hot; and, as
the service was entirely in Gaelic, my attention was wholly
drawn to the gulls, whose instincts had brought them where
they would find an ample supply of insect food. They flew
to and fro, snapping up the flies with amusing dexterity ; and
though I did not remain longer than an hour at the place,
I was afterwards informed they had continued their soaring
flight until the assemblage dispersed.

POMERINE SKUA (Lestris pomerinus). I have recognized
this bird off Brodick, and also in Kilbrannan Sound.

RICHARDSON'S SKUA (Lestris Ricliardsonii). This pirate
bird has also been recognized in passing during the autumn
months.

MANX SHEARWATER (Puffinus anglorum). Recognized by
myself off Brodick, on more than one occasion.

FORK-TAILED PETREL (Thalassidroma Leachii). Occurs in
the Firth of Clyde every winter, and is occasionally blown on
shore during a storm of unusual severity.

STORM PETREL (Thalassidroma pelagica). I have seen
small parties of this interesting little bird on dull days while
boating in the firth. A few pairs breed on Ailsa Craig, where
I have seen them quitting their holes at mid-day.

312 BIRDS OF ARRAN.

129. CONCLUDING REMARKS.

It will be seen from a glance at the preceding list that it
does not contain much that may be said to possess special
interest. Compared with the counties of Ayr and Wigtown,
the islands forming Buteshire present a smaller number of
species. A patient research, especially if carried out by
some one resident on the islands, might result in very many
additions to the present catalogue. Meanwhile, the chief
interest centres in the more conspicuous birds of prey; and it
is particularly gratifying to know that these are now to be
protected. As I have stated in the notes on the golden
eagle, hopeful signs are already apparent, two eyries of that
bird having this year been found within forty miles of the
city of Glasgow. A better refuge than Arran for eagles
there could not be; and should the now deserted haunts
be again occupied, I trust that no west country landowner
will allow his keeper to kill the poor birds should they at
any time venture beyond the boundaries of their own hunting
grounds.

313

THE MAMMALS OF AKHAN.

130. The Mammals of Arran are few in number. As far
as I can learn, only fourteen species are found in the island,
and of these, two (the seal and porpoise) are natives of the
sea, while four (the rat, mouse, hare, and rabbit) have been
introduced by man ; thus leaving only eight species of native
quadrupeds. The mole and hedgehog are absent, as are all
the carnivora, save the otter; even the weasel, which is
found in the neighbouring island of Bute, is a stranger to
Arran. The wild cat is mentioned as a native of this island
by Pennant in his Tour in Scotland, but it is now un-
known, and it seems probable that he was misinformed, and
that the animals alluded to were merely house cats run wild.
The true wild cat does not appear to be found in any of the
Scotch islands.

One interesting animal must be mentioned, although it
has, of course, no claim to be regarded as a member of our
fauna. This is the Virginian Deer (Cervus Virginianus,
GMEL.) of America, which was introduced into Arran about
the year 1832. They soon became very plentiful, but of
late years they have been considerably reduced in numbers.
They seem to thrive well, but their antlers are not so large
as American specimens.

In drawing up the following list, I have to acknowledge
the kind help of Mr. Halliday, the Duke of Hamilton's
head-keeper in Arran, from whom I have received much
information.

1. Common Bat ( Vesperugo pipistreUus, SCHREB.) Com-
mon.

314 MAMMALS OF ARRAN.

2. Long-eared Bat (Plecotus auritus, LINN.) Common.

3. Common Shrew (Sorex vulgaris, LINN.) Common.

4. Black Water-Shrew (Crossopus fodiens, PALL.) The dark
race of the water-shrew (often, described as a distinct species
under the name of C. remifer) is not uncommon in the burns
near Loch Ranza. This variety is pretty widely spread
throughout Scotland, while the white-breasted race seems to
be very rare or local Specimens intermediate in character
are often taken.

5. Otter (Lutra vulgaris, ERXL.) Common. Except tame
cats run wild, and rats, this is the only sort of " ground
vermin " found in Arran.

6. Common Seal (Phoca vitulina, LINN.) Sometimes seen,
but is not plentiful on any part of the coast

7. Porpoise (Phoccena communis, F. Cuv.) Frequently
seen in pursuit of the shoals of herrings. Other species of
cetaceans may perhaps visit the coast, but I am not aware
that any have yet been recorded.

8. Brown Rat (Mus decumanus, PALL.) Abundant, both
in houses and in the open country. Mr. Halliday observes
that, when hunted, it takes to water as freely as the true
water-vole.

9. House Mouse (Mus musculus, LINN.) Common.

10. Wood Mouse (Mus sylvaticus, LINN.) Common.

11. Common Field-vole (Arvicola agrestis, LINN.) Com-
mon. The red field- vole (A. glareolus, SCHREB.) may pos-
sibly be also found, for it is widely distributed in Scotland.
Mr. Halliday has never seen the water-vole (A. amphibius,
LINN.) in Arran, and believes that it is not a native of the
island.

12. Common Hare (Lepus timidus, LINN.) Common in
suitable places, but is believed to have been introduced from
the mainland. It is not mentioned by Pennant.

13. Rabbit (Lepus cuniculua, LINN.) Rabbits must have
been introduced before 1772; for they are included by
Pennant among the quadrupeds of the island.

MAMMALS OF ARRAN. 315

14. Red Deer (Cervus elaphus, LINN.) Martin, in his
curious Description of the Western Islands, writes of
Arran in the beginning of the last century : — " The highest
Hills of this Island are seen at a considerable distance from
several parts of the Continent and North-west Isles, and they
serve instead of a forest to maintain the deer, which are
about four hundred in number, and they are carefully kept
by a forester — to give sport to the Duke of Hamilton, or
any of his family, that go a hunting there. For, if any of
the natives happen to kill a deer without license, which is
not often granted, he is liable to a fine of <£20 Scots for each
deer. And when they grow too numerous, the forester grants
licenses for killing a certain number of them, on condition
they bring the skins to himself." After this they became
scarcer, for Pennant says, in 1772, that "the stags, which
used to abound, are now reduced to about a dozen." The
original breed was never exterminated, however, and in
February, 1859, an importation of fresh blood from the
mainland was made, with the best effect. Mr. Halliday now
estimates the number of red deer in the island at about five
hundred head. Being little disturbed, save in the season,
they frequent the woods and cultivated ground more than
is usually the case in Scotland; and in consequence, doubtless
of the good feeding, their horns are often very fine. A stag,
shot by the Duke of Hamilton, in 1867, weighed 22 imperial
stones, and carried a very symmetrical head of fourteen points;
another with fifteen tynes was killed in the same year. The
roe appears never to have been found in Arran.

316

GEOLOGT OF BUTE.

131. A PRETTY full and generally accurate account of the
Geology of Bute has been given by Dr. MacCulloch in his
work on the Western Isles, published in 1819. For thirty
years after, no observations, so far as we can learn, were
put on record respecting it, except some notices of its
remarkable coast terrace and raised shelly deposits, by Mr.
Smith of Jordanhill. The greater variety of the strata in
Arran, and the bearing of the phenomena there exhibited
upon questions in theoretical geology actively discussed at
the time, drew attention entirely to that island; and Bute,
in common with other parts of the west of Scotland, was
overlooked. Yet it has many points of great interest;
strata occur here to be m et with nowhere else in Scotland,
and the effects of its trap dikes upon the adjoining strata are
extremely curious — indeed, of a unique character.

Physical Features.

132. Three deep depressions or valleys traverse Bute
perpendicularly to its longer axis, dividing the island into
four portions, and marking the boundaries of distinct
geological formations. They terminate on either side in
bays or indentations of the land, formed here, as in most
other cases, at the points of least resistance, the junc-
tions, namely, of dissimilar strata. Those on the east side
are the well-known sheltered bays of Kames, Rothesay, and
Kilchattan. The low tracts in question shew no rock in
situ, but are filled with shingle and alluvial deposits conceal-
ing the junctions, strata of peat, and occasional shell-beds.
The elevation above the sea level nowhere exceeds 30 feet ;
and as this is also very nearly the height of the terrace

GEOLOGY OF BUTE. 317

already referred to as encircling the island, it appears that
when the sea stood at that ancient level, Bute consisted of
four islands, separated by narrow channels.

Slates and Sandstone.

The various strata exhibited in Bute are the terminal
portions of those great bands of rock, sedimentary and
igneous, which extend across the country from sea to sea, as
already noticed. Mica slate occupies the northern portion,
between the Kyles on the north and Kames and Ettrick
Bays on the south. The rock has its usual character and

Fig. 37.

a, Kames bay; b, Rothesay ; c, Kilchattan; m, mica slate; n, clay and
chlorite slates; s s, red sandstone; t, trap; r, the terrace.

aspect, and rises into hills of nearly 1000 feet elevation.
The district south of this, bounded by the Rothesay valley,
consists of the two upper slates, the common clay, and
chloritic. Subordinate to these are great beds of quartz
rock, the most considerable of which forms the high ridge
called Barone Hill, with a picturesque old ruin overlooking
Rothesay on the west. There are also copper veins in this
slate westwards from Kames Bay. The portion extending
from Rothesay Valley to Kilchattau is occupied by red sand-
stone; and finally the southern portion, with a substratum
of red sandstone, consists mainly of various rocks of the trap
family, erupted through and overlying the sandstone. The
accompanying outline of the island shews the relation of
these strata to the valleys or depressions, which are obviously
a part of the system of parallel fractures ranging N.E. and
S.W. across Scotland, and probably due to the upheaval of
the mountains of the Central Highlands.

A description of these rocks would be useless, as it would

318 GEOLOGY OP BUTE.

merely be a repetition of facts contained in all elementary
works. The sandstone is in the place of the Old Red,
as succeeding the slates, and has much of the mineral
character of this formation in other districts. Along the
E. shore in several places, and again at Kilchattan, beds of
limestone occur subordinate to it; but unlike the Arran
limestones in the same subordinate positions, they have no
fossils; containing much siliceous matter, they are of little
economic value, and thus have many analogies with the corn-
stones of the Old Red, such as occur near the base of Ben-
Lomond and in several localities in Arran, the march of
Achab farm, the Fallen Rocks, Glen Loig, and Kilchattan.
It is very doubtful, however, whether the rock is really of
this age.*

Carboniferous Beds.

133. But there are in Bute sandstones and limestones
newer than these, which have been till very recently quite
overlooked by geologists. There occurs, in fact, at Ascog a
small coal formation which may yet turn out to be of
economic value. It is connected with an isolated, overlying
mass of trap, appearing on the shore and occupying the cliffs
near Ascog mill. On the north side of the promontory,
south of the mill, several thin courses of nodular limestone
traverse beds of brown, crumbling shale, subordinate to sand-
stone. The shale is of considerable thickness, and rises into
banks above the road. The south side of the promontory
presents the following section (fig. 38).

The lowest bed, o, is a fine-grained, bluish-gray, nodular

• In the last edition, the lowest beds of the sandstone were referred
to the Old Bed ; a late visit to Bute, however, led me to doubt this
view, and the qualified statement in the text was then made. I have
since learned that Professor Geikie considers all the sandstones to be
newer than the Old Bed. It seems to me, nevertheless, to be highly
probable that the lowest bands traversing the obscured belt of low
land along the border of the slate may really yet be found to be of
this age.

GEOLOGY OF BUTE.

319

limestone. Over it is a bed of black bituminous shale, b,
containing veins of coal about a quarter of an inch thick ;
and upon these rests a bed of concretipnary limestone, c, the
base or paste being a dark-coloured limestone, and the
concretions rounded lumps of the same rock, often of con-
siderable size. The upper part of the cliff is occupied
by trap in various prismatic forms. The base of the con-
cretionary limestone is so much altered by the contact of
the trap, that the two rocks can only be distinguished
by the action of a strong acid. A like change is produced
upon the imbedded lumps in the upper part of the bed. The
limestone, shale, and coal seams extend under high-water

Fig. 38.

a, Limestone; 6, shale with thin coal seams; c, b'mestone breccia; d,

trap.

mark, and when the tide is very low considerable pieces
of coal are often dug out from beneath the sand and mud
covering the tide-way. Several sinkings have been made
here for the purpose of discovering workable coal seams,
but without success. The Eev. A. M'Bride, of Port Banna-
tyne, informs me that stigmaria and other fossils of the coal
formation have been found by him in the sandstone here,
thus bringing it into close analogy with the Corrie beds.

620 GEOLOGY OF BUTE.

The Ascog coal measures may, however, be but a local deposit
laid over the older beds, and may not extend further along
the coast; and it is not improbable that the lower band only
of the sandstone next the slate may be of the Old Red forma-
tion, while the outer portions may correspond to the con-
glomerates of Brodick Bay. The strata are not exposed in
deep glens and high mountain sides, as in Arran ; and the
exploration of the beds is thus a matter of much greater
difficulty.

Trappean Beds.

134. The trap above referred to is a projection from the
principal mass above, which occupies a considerable area
inland towards Ascog Lake, and is upwards of 100 feet thick.
It is seen in the summit of the cliffs on either side, the line
of junction ascending rapidly as it retires from the shore.
These trap rocks at Ascog derive their chief interest from
being the repository of beds of lignite — a substance so rare
in Scotland, that no well-marked beds occur on the main-
land, and but few in the other islands; and these in situa-
tions very difficult of access.

The principal bed is situated in the face of the cliffs above
the road, a little to the south of Ascog mill, as shewn in the
annexed section (fig. 39), giving the various beds.

The lowest bed resting on the sandstone is a small-grained,
rudely columnar greenstone; the junction is, however, con-
cealed. Over this is a trap-tuff with a base of greenstone,
and imbedded spherical lumps of the same substance. This
is followed by a bed of red ochre of coarse texture, traversed
by numerous black iron seams, which have doubtless been
produced from a change in the oxidation of the component
iron. Over this is the lignite bed : it is three feet thick, and
consists of hard stony coal, interstratified with a yellowish-
white shale, both being much intermixed with pyrites. The
coal has been so much altered throughout its whole thickness,
by the contact of the trap rock, that Mr. Rose of Edinburgh,

GEOLOGY OF BUTE.

321

to whose examination the best specimens that could be
selected were submitted, in order that he might determine

Fig. 39.

s, Sandstone; r, terrace and road;//, greenstone; a, trap-tuff; 6, red
ochre ; c, lignite bed ; d, pisolitic ochre ; e, porphyritic amygdaloid, the
upper portion much altered.

the species of wood, but without any note of the geological
situation of the coal, was "unable to obtain a slice, in conse-
quence of the structure being altered by the contact of a
whin dike." The coal has been worked to some extent
by driving an adit inwards on the line of the dip, which is
about 20° to the westward ; but the workings have been for
some time abandoned, and the inner and lower portions are
now full of water.

The floor of the coal has been already described. The
roof is a peculiar rock. It consists of a base or paste of an
ochreous steatite, with imbedded round pieces of the same
substance, and may hence be called a pisolitic ochre; it is
three and a-half yards thick. The bed above this is of the
same character; but the base feels less unctuous, and with
the imbedded steatite it contains also imbedded calcareous
spar. The base effervesces briskly with an acid ; and hence
we may call the rock a calcareous amygdaloid. The upper
portion of this bed, to the thickness of a few inches only, is

322 GEOLOGY OF BUTE.

very hard, and has a semivitreous appearance, and thus
closely resembles a porphyry. In common with the trap
above, and indeed all the beds in this locality, it contains
much disseminated iron. The rest of the cliff is occupied
by common greenstone or diorite, similar to the lower bed
in contact with the sandstone.

Another bed of lignite occurs on the opposite, or north-
west side of the trap district, overlooking Ascog Lake. The
coal dips to the interior of the area — that is, nearly south.
It is of about the same thickness, and is accompanied by
beds of steatite and red ochre, very similar to those above
described ; but the nature of the ground is such that a com-
plete section cannot be had, and the precise number, there-
fore, and order of the beds, cannot be exactly stated. The asso-
ciation, however, of the lignite with ochres and steatites here
also is sufficiently distinct; and it is even probable that these
beds are persistent throughout the whole of this district.

135. In Mull, Skye, and some others of the western isles, and
in the basaltic district of the north-east of Ireland, lignites
occur in the middle and upper portions of the series, asso-
ciated with variegated ochre. The leaf-beds of Antrim,
associated with ochre and lignite, discovered by us in 1844,
and those of Mull in like association, discovered by the
Duke of Argyll in 1850 (without knowledge of our dis-
covery), contain the same species of plants. This similarity
and that of the beds clearly shew that similar condi-
tions have prevailed over a very wide area, the suc-
cessive eruptions of igneous matter over the sea bottom
being very similar, and that there were like periods of
repose, during which the productions of the adjoining land
were swept down to be buried under the next flow of sub-
marine lava.

Now, the basaltic series of the north-east of Ireland, as it
overlies the chalk formation, clearly belongs to the tertiary
era, and was long ago recognized as of this age. But many
cases occur in which the same basaltic flow, which alters the

GEOLOGY OF BUTE.

323

chalk to the state of a saccharine marble, spreads out beyond
the limits of the chalk, and overlies and alters the new red
sandstone and coal measures, just as traps rising through the
Old Red in Clydesdale overflow upon the coal formation.
Such overlying masses must also belong to the tertiary
period: and the distribution of this basaltic series over the
western isles, in direct continuation of those of north-east
Ireland, long ago led to the conclusion that they were of the
same age with them, and that this vast area, ranging from
Skye to Londonderry, had been once the scene of contempor-
aneous volcanic activity. A step in the wrong direction,
was taken by Edward Forbes, when mistaking the structure
of Trotternish, he endeavoured to shew that the traps there
were of oolitic age. This mistake was several years ago
pointed out by us, and the deceptive appearances explained.
It is now, and we think rightly, agreed, mainly from the
evidence of the lignites and leaf-beds, that the entire series
belongs to the middle tertiary — that is, miocene period.
Conybeare and Buckland, half a century ago, referred the
traps of north-east Ireland and the Scottish Isles to one
era of eruption, of later date than the chalk ; and it is not
improbable that these Bute lignites, ochres, and traps, though
overlying carboniferous beds, may really be of the same tertiary
age. The wood is so much altered that we have failed to
determine its true character; but as Professors Harkness
and Blyth, by microscopic and chemical examination, have
shewn the identity of the Mull and Antrim lignites (Ed.
Phil. Jour. N.S., vol. iv. p. 304, 1856), there can be little doubt
that those of Bute are the same — and so presumably of the
same age.

136. The chemical composition of ochre is almost exactly
the same as that of basalt: 100 parts consists of — silex, 5640 ;
alumina, 3'46 ; per-ox. iron, 24-14; carb. lime, 0'90 ; water,
15 -10. It is, in fact, a decomposed basalt, or volcanic ash,
pai-tially re-fused and reconsolidated by the succeeding flow
of igneous matter.

324

GEOLOGY OF BUTE.

Limestone.

137. The limestone of Kilchattan Bay is subordinate to
the sandstone; the strata of the two rocks are conformable,
and the dip nearly south at a moderate angle. At the
summit of the ridge, near the picturesque ruins of the
ancient castle of Kelspoke, the sandstone over the limestone
is seen dipping under the trap which bounds the rugged
terraced ridges, descending towards Garrochhead on the
south coast. These ridges have the same inclination south-
wards as the underlying sandstone strata, and present a

Fig. 40.
a,b, Sandstone; c, limestone; rl, trap.

succession of bold fronts towards the north. The sandstone
appears to have had its present inclination when the sub-
marine lava streams, of which these ridges consist, were
poured out over it; the scarped fronts were no doubt formed
by the action of currents when the land was rising. The
arrangement of the strata at Kilchattan is shewn in the
annexed cut (fig. 40). No fossils were found in the limestone.

Dikes and tJieir Effects.

138. The dikes of Bute are composed of greenstone or
basalt, and are very numerous, especially on the east coast.
They traverse the strata in various directions, and in some
cases can be traced for several miles continuously, preserving
nearly the same width and direction throughout. Two or
more are sometimes seen to meet and coalesce for some
distance, and again to separate; a narrow dike branches

GEOLOGY OF BUTE. 325

off in several filaments, which unite again — portions of
the rock traversed are often found entangled in the dike;
and these, as well as the contiguous strata, present the usual
metamorphic effects recognized as due to igneous action, be-
sides others of a peculiar and exceptional character, to which
we shall presently allude.

A dike seventy feet wide emerges from the sea at the
mineral well near Bogany Point, and ranging nearly west
is seen in Huntly Place, where it has been largely quarried;
interrupted by the. bay, it rises again, is conspicuous across
the high grounds west of the town, and, crossing the island,
enters the sea at Ettrick Bay. Here another dike, two or
three times the width, enters near it, crossing from Ascog,
and visible in several eminences in the interior. The Rothe-
say mineral water, which has acquired some reputation for
efficacy in rheumatic, cutaneous, and glandular complaints,
rises in the former of these two dikes at Bogany Point. The
gallon of 277*274 cub. inches contains, according to the
analysis of the late Dr. Thomas Thomson, Professor of
Chemistry in Glasgow University, — Com. salt, 1860*73
grains ; sulph. lime, 125-20 ; sulph. soda, 129-77 ; mur.
magnesia, 32-80; silica, 14-39: sulphuretted hydrogen, 17-4
cub. inches. Both dikes and this mineral water are noticed
at some length in "Wilson's Guide to Rothesay and the Island
of Bute, an excellent and neatly illustrated little work.

The Kilchattan limestone is altered in a remarkable manner
by a large dike, crossing it nearly in the direction of the dip,
and the effects are well seen at the eastern side of the quarry.
Along the plane of contact with the dike the limestone is
altered to the state of a granular saccharine marble, which
on the application of a slight pressure crumbles into a fine
powder. This is succeeded by a hard crystalline marble, the
crystals appearing in distinct flakes. Between this and the
first change, which is one of simple induration, there are
many gradations. Similar effects are common at the con-
tact of limestone with plutonic rocks ; in some localities they

326 GEOLOGY OF BUTE.

are accompanied by other singular changes of a chemical
nature. Magnesia, and sometimes silica and alumina, are
introduced into the composition of the limestone, so that
simple carbonate of lime becomes a double carbonate of lime
and magnesia. The source whence this magnesia has been
derived has occasioned much difference of opinion among
geologists. Some imagine that it has been transferred from
the plutonic rock to the limestone; while others hold that,
as fractures and dislocations of the earth's crust accompanied
the eruption of these plutonic rocks, gaseous exhalations
might find their way from beneath, and introduce carbonate
of magnesia and other substances into rocks near the surface.
In confirmation of this view, Mr. Phillips has shewn, in his
Geology of Yorkshire, that " common limestone is dolomitized
•by the sides of faults and mineral veins far away from igneous
rocks of any kind ;" and some distinguished chemists have
expressed their belief that carbonate of magnesia may be
sublimed by the action of great heat. (Rep. Brit. Assoc. for
1835, Trans. Sect, p. 51 ; Philips's Geology, in Cab. Cyclop.,
vol. ii. p. 98.) Much doubt, however, still hangs about
this subject. Cases occur in which magnesia has been in-
troduced, although the limestone could not have been subject
to such a pressure as would confine its carbonic acid when
the rock was softened by heat.

In order to elucidate, if possible, this obscure subject, two
specimens of the rock were submitted to the late Professor
Macadam, at that time lecturer on chemistry in Edinburgh,
afterwards Professor in Melbourne College, for examination
with reference to the presence or absence of magnesia.
Specimen No. 1 is the saccharine marble from contact with
the dike; No 2 is the unaltered limestone — both average
specimens : —

" In specimen No. 1, carbonate of magnesia constitutes
about 2£ per cent, of the whole mass. Its other and
principal ingredients are carbonic acid and lime, silica, and
traces of oxide of iron and alumina.

GEOLOGY OF BUTE. 327

" In specimen No. 2, magnesia abounds, the amount pre-
sent being equivalent to 33 '72 per cent, of carbonate of
magnesia. The other constituents are similar to those in.
No. 1. From the large proportions of carbonate of lime and
carbonate of magnesia, No. 2 would appear to be a species
of dolomite. It may be noticed that the physical characters
of No. 2 are very different from those of No. 1 ; the former
is difficult to pulverize, the latter is extremely susceptible of
division.

" The action of strong hydrochloric acid on both specimens
causes a portion of gelatinous silica to appear, shewing the
presence of a silicate, which may be that of magnesia, since
the quantity of gelatinous silica is about sufficient to combine
with the 1'28 per cent, of caustic magnesia existing in the
specimen No. 1. There is a less quantity of this gelatinous
silica in No. 2. The greater portion, however, of the silica
present in both specimens remains undissolved in the gritty
or pulverulent condition, and is hence in a state of mere
mechanical mixture with the other constituents of the lime-
stone. It would require a minute quantitative analysis to
determine whether the 1 *28 per cent, of magnesia exists as a
carbonate or silicate, or partly as both."

The phenomena are thus of a contrary character to what
is usually found; the unaltered rock is a dolomite, and con-
tains nearly 34 per cent, of carbonate of magnesia, while the
altered rock contains less than 3 per cent. What has
become of the constituent magnesia? Has it been driven
off by the heat to which the limestone was exposed? Most
chemists are unwilling to admit that this is possible; and it
may reasonably be objected, that if the limestone had been
exposed to so high a temperature as to vapourize its magnesia,
the silica would not be mechanically present, but would have
entered into chemical combination with the lime or magnesia,
and have formed a silicate.

That whin dikes have sometimes been the means of pro-
ducing such a combination has been shewn by an eminent

328 GEOLOGY OF BUTE.

chemist. Dr. Apjohn found the white chalk of Antrim,
altered by whin dikes to the state of a saccharine marble, to
be a trisilicate of lime, " very analogous in its composition
to olivine. We are thus enabled to understand why olivine
should be so very frequently found in trap rocks, and to
refer its origin to the contact of silex at a high temperature
with an excess of the basic oxides; and we have in some
degree a demonstration that the dolomites which contain
siliceous sand could not have been exposed at any time to a
heat sufficiently high to account for the introduction into
them of magnesia in the vaporous state; for by such a heat a
silicate of lime or magnesia, or of both, would have been pro-
duced " (Jour. Geol. Soc. Dub., vol. i. p. 376).

The presence of these silicates in both our specimens is
shewn by the appearance of the gelatinous silica; yet a
greater quantity of silica is present mechanically, which, as
already stated, seems inconsistent with the exposure of the
rock to intense heat; unless, indeed, we could suppose that
the silica has been introduced by infiltration, or the magnesia
removed by the solvent power of free carbonic acid at a period
subsequent to the consolidation of the dike from a state of
igneous fusion.

139. Careful quantitative analyses of the limestones were
made by the late Dr. Robert D. Thomson. It is hoped that
these will afford definite terms of comparison with other
analyses, such as those of Dr. Apjohn, already referred to,
and that their publication may lead to the formation of
clearer views respecting an obscure question in theoretical
geology.

The analyses are as follow : —

Specimen No. 1 is the saccharine marble from contact with
the dike at Kilchattan — in the highest state of alteration.

No. 2 is the hard crystalline marble, having the crystals in
distinct flakes, more remote and less altered than No. 1.

No. 3 is the unaltered limestone from the middle of the
quarry, remote from the dike — an average specimen.

GEOLOGY OF BUTE. 329

No. 4 is the altered limestoue from contact with the over-
lying trap at Ascog Mill; it is an impure, dark-coloured
rock, of an earthy aspect, and very like the trap which rests
upon it.

No. 1.— Spec. grav. 2-710.

I 6-91 5-16 5-70

J

.
Alumina,

Protoxide of iron, ................ 1-68 1-50 1'28

Carbonate of lime, ............... 90-65 91-08

Carbonate of magnesia, ......... 1-00 1-17

No. 3.— Spec. grav. 2-679.

100-24 99-23

No. 2. — Spec. grav. 2-570.

I. II.

1 1-94 0-28 0-28

Alumina, /

Protoxide of iron, 0-52 0-56

Carbonate of lime, 96-48 98-76 96-58

Carbonate of magnesia, 1-23 — 2-24

100-17 99-66

Protoxide of iron, ............... 1-12 1-12

Carbonate of lime, ............... 67'42 72-12 67-00

Carbonate of magnesia, ......... 17'31 — 18-06

Water, coalv matter, and )

Carbonic acid, ...... ! ...... } 4>45 4'74

100-00 100-00

No. 4.

Protoxide of iron, ............... 6-42 6-60

Carbonate of lime, ............... 24-00 21-20

Carbonate of magnesia, ......... 4-62 2'8o

Water and carbonic acid, ...... 1 -75 4-89

100-00 100-00

330 GEOLOGY OF BUTE.

The silica present is iu a state of mechanical mixture.

These analyses are confirmatory of the main points of the
views already stated, and seem clearly to establish the new
and remarkable fact, that by the igneous action in these
instances the magnesia has been driven off from the lime-
stone. The unaltered rock is a dolomite containing nearly
70 per cent, of carbonate of lime, and nearly 20 per
cent, of carbonate of magnesia; while the altered rock
contains but from 1 to 2 per cent, of he latter in-
gredient. To what cause are we to assign the changes that
have taken place? Has the magnesia been sublimed by
heat? or has it been withdrawn by the solvent power of
free carbonic acid? On the nature of these and the other
chemical changes that have been induced, it is difficult to
express an opinion; and from such limited premises it
would be unphilosophical to draw any general conclusions.
The subject is one, however, of great interest both to the
geologist and chemist, as the facts are directly opposed to
the received views ; and as no instance of similar changes on
dolomitic rocks has, so far as we are aware, ever been put on
record.

140. A remarkable dike intersects the sandstone between
Ascog and Kerrycroy, well worthy of attention on account of
the striking illustration it affords of the mode of cooling of
basaltic rocks from igneous fusion. It runs parallel to the
shore for some distance, and then retires from it towards
the south-west, and striking the inland cliff already mentioned,
whose direction is here the same as that of the dike, it forms
the perpendicular face of the cliff in front of the sandstone,
rising like a wall to the height of twenty or thirty feet. The
direction of the cliff soon changes, however, and the dike
then enters the hill behind, and is lost. The sandstone
having been completely worn away from the seaward face
of the dike, a very large surface of the side is laid bare, and
thus the structure is well exhibited, as in the annexed cut
(li*. 41).

GEOLOGY OP BUTE. 331

The dike is composed of greenstone, is about fifteen feet
wide, and the prisms are mostly pentagons and hexagons.

Fig. 41.
Side view of whin dike between Ascog and Kerrycroy.

The cause of this peculiar structure has been already ex-
plained to be the parting of the heat of fusion laterally, per-
pendicular to the cooling surfaces, and the development
therefore of the prismatic structure in a horizontal direction.

The Arctic Shell Beds.

141. The north shore of Bute is classic ground in the history
of the glacial beds of Scotland. It was from an examination
of the shell beds found here that Mr. Smith of Jordanhill
was first led to recognize the arctic character of these deposits.
At Balnacaillie Bay, in the Kyles, opposite the Burnt Isles,
the beds occur first explored by him, still one of the richest
localities in the basin of the Clyde. A stream on the Bute
side of the Kyle has cut out the bed and exposed the shells;
the bed extends a considerable way up a hollow and along
the level shore. In such a locality great care must be taken
to distinguish the shells of the ancient beds from those of the
present sea, as well as those of one bed from those of another ;
since those successively newer overlap the older as we advance

332 GEOLOGY OF BUTE.

seaward. The cautions given in Arts. 85 and 90 should also
be attended to in examining these beds. In this locality
Mr. Smith found the following species : —

Cyprina Islandica. Lacuna vincta.

Astarte Garensis. Modiola modiolus.

- elliptica. Fusus antiquus.

- unicostata. Nucula nucleus.
Tellina proxima. Leda oblonga.
Pecten Islandicus. Saxicava rugosa.

— opercularis. sulcata.

Natica clausa. Panopjea Norvegica,

glaucinoides. Mya udevallensis.

Balanus costatus. — truncata.

Serpula triquetra. arenaria.

Spirorbis corrugatus. Cardium edula

Littorina littorea. Sphenia Suainsoni.*

To these Mr. Crosskey added the following: —

Natica Groenlandica. Margarita unduluta.
Mangelia pyi-amidalis. Crenella nigra.
var. rosea (Sars).

This is the best locality for the Panopcea Norvegica and
Pecten Islandicus; the latter occurs in large quantities, and
perfect specimens are found with northern Balanus attached.
One valve of Crenella was found. The boring shetl, Mya
udevallensis, occurs in the upright boring position in great
numbers.

Among those who studied the glacial beds wnce Mr. Smith
ceased to work entirely among them, the Rev. A. MacBride,
of Ardmory, Bute, the Rev. H. W. Crosskey, now resident
in Birmingham, and Mr. David Robertson have been the

* The names are here given according to the original authorities.
< 'onchologists will at once perceive that the nomenclature is not
uniform.

GEOLOGY OF BUTE. 333

most zealous and successful. The Rev. Dr. Frazer, Paisley,
and Mr. Craig, Beith, have also done eminent service in the
same field. Mr. MacBride was the first to publish a division
of the beds, applicable to Bute, which has been since extended
by various researches, and the division may be considered to
hold true for all the beds in the west of Scotland. There
are local variations, but the division established by Mr.
MacBride represents the normal development of the series,
not only in Clydesdale, but generally over Scotland. The
Arran beds, we have seen, form one of the exceptions. The
series is shewn in its complete development at Kilchattau
Bay, close to the tile works. Mr. MacBride's order of suc-
cession (descending) is as follows : —

1. Vegetable soil.

2. Sand and gravel, stratified and passing downwards into
a sandy clay, with gravel, 10 to 12 feet.

3. Red clay, olive-green below, without stones or shells,
1 to 2 feet.

4. Fine dark clay with arctic shells, 2 feet.

5. Fine laminated clay, red or brown, without shells or
stones, 15 to 18 feet.

6. Boulder clay, with striated stones; the upper surface
hummocky and irregular.

Here bed No. 5 is the bed absent in Arran, but usually
present all over Clydesdale. It is a fine fissile clay, easily
opening into laminae like the leaves of a book, without
stones, and considered to be entirely unfossiliferous till the
recent discovery in it of foraminifera by Rev. H. "W. Cross-
key and Mr. D. Robertson; it is thus remarkably contrasted
with the boulder clay on which it rests. There are also here
indications of an upper shell bed. This section is contrasted
with those we have given in Arran (Art. 86), by the presence
of the laminated clay, always absent there, so that the bed
which is the main depository of the arctic shells rests directly
on the Till. Throughout Bute, on the opposite coast of Cowal
and Renfrew, and in most parts of Clydesdale, this laminated

334 GEOLOGY OF BUTE.

clay is always interposed, and is a remarkable feature in the
beds. It is very striking to pass at once from this clay,
mostly unfossiliferous and without stones, to a rich shell bed
in which stones also are contained. The absence from Arran
of this laminated bed may perhaps be accounted for by the
exposure of the situation ; for if it was formed by the trans-
port of fine mud from the ends of glaciers which reached
the coast of the period, — and this seems the most probable
origin for it — then on situations more open and exposed it
would be swept away.

List of Species.

142. These fossiliferous clays may be seen at many places on
the Bute coast, besides the two chief localities we have named —
as Kames Bay, Ettrick Bay, where they are now nearly
denuded by the sea; parts of Eothesay Bay, and indeed
wherever there are sheltered places suited to their preserva-
tion. They have been long carefully studied by the Rev. A.
MacBride. Later and more extended lists are given in a
series of papers, not yet completed, by Rev. H. W. Crosskey
and Mr. D. Robertson, in the Transactions of the Geological
Society, Glasgow. The following list of the species found at
Kilchattan has been kindly suppliedx to me by Mr. D.
Robertson. It contains all the fossils from these beds known
to him, except those of the Ostracoda; on these a monograph
is now in preparation : —

Fislws.
A few vertebrae.

Conchifera.

Anomia ephippium, .... Lin.

Pecten Islandicus, .... Mull.

My til us modiolus, .... Lin.

Nercula tennis, Mont.

Leda pernula, var., ....

GEOLOGY OF BUTE.

335

Leda mucilenta,

pygmsea, var.,

lenticula, .

Axinus flexuosus, var., Gouldii,
Montacuta ferruginosa,
Cardiuin edule, .

— echinatum, .

Cyprina Islandica,
Astarte compressa,
Tellina calcarea,
Mactra subtruncata, .
Scrobicularia prismatica,
Mya truncata, .
Pholas crispata, .
Tectura virginea,
Trochus cinerarias,

• tumidus,

Lacuna divaricata,
Littorina littorea,

obtusata,

rudis, .

Rissoa parva, var., interruputa,

striata,

Skenia planorbis,
Homalogyra atom us, .
Odostoma (?),
Natica Groenlandica, .

• affinis,

Velutina Isevigata, . - «
Aporrhais pes-pelicani,
Purpura lapillus,
Buccinum undatum, . .
Trophou clathratus,

tntncatus, . .

Fusus antiquixs, .
Pleurotoma violacea, .

turricula, .

Mangelia pyramidalis,
Totriculas obtusus, . • .

Steenst.
Moll.

Mont.

Lin.

Do.

Do.

Mont.

Chem.

Mont.

Do.

Lin.

Do.

Mull.

Lin.

Do.

Fabr.

Lin.

Do.

Manton.

Adams.

Fabr.

Philippi.

Beck.

Ginel.

Lin.

Dacosta.

Lin.

Do.

F. &H.

Strom.

Lin.

Megb. & Ad.

Mont.

Mont.

Crustacea,
Claws and plates of crabs.

33G GEOLOGY OF BUTE.

Cirripoda,

Balanus creiiatus,

porcatus, . . . .

Verruca stromia, .... Miill.

Annelida.
Serpula verruicularis, .... Lin.

Echinodermata.

Echinus sphsvra, plates and spines,

• drobachiensis, . . . Miill.

Striated Rocks.

143. Striated and polished rocks, striated boulders, and
roches moutonnees are plentifully met with in Bute ; the whole
contour, indeed, of the surface in all its three divisions clearly
points to ice action as having moulded the outlines. Perhaps
one of the most perfect specimens of striation and rounding to
be met with in Scotland occurs on the high ground W. of
Rothesay. The slate and quartz rock on and in the neighbour-
hood of Barone Hill, 520 feet in height, have been smoothed,
polished, and striated in the most i-emarkable way by a force
acting from the north up tli& slope on that side and down on the
other, the nigged prominences of the strata which rise toward
the north being worn off, while the surfaces directed the
opposite way retain their original outlines. Mr. Geikie
(Tr. GeoL Soc. GL, L 23) gives another striking example,
not visited by us, on the shore of Loch Dhu; and the round-
ing and polishing on the N. end of the island and in the
Kyles generally can hardly have escaped the notice of any
one who has made the passage of that lovely strait on a
bright day. Mr. MacBride, by whom most parts of the
island have been carefully examined, with a view to the
evidences of glacial action, considers that the abrading agent
cume from the N., and from without the island, and passing
steadily over every part in one uniform direction, save with

GEOLOGY OF BUTE. 337

certain trifling local deflections, passed out into the channel
which divides Bute from Arran. He holds strongly the
view that the granite boulders one sees along the S. and
S. W. shores of Bute are not from Arran, but are ship ballast
tossed overboard and washed up. He maintains also that
the statement recently put forth, that Arran granites are
found on the Cantire coast, is certainly erroneous ; and that
a northern or north-eastern origin must be sought for them.
We have not studied the island so carefully as to pronounce
any positive opinion on this point; but certainly there is
no extensive elevated surface in Bute to which we could
look as a snowfield, or birth-place of glaciers.

Features of the

144. The facts above stated point to a considerable
and extensive depression of the laud, as the highest
beds with shells reach 510 feet, and are as far inland as
Airdrie; yet this amount of depression would preserve all the
dominant features of the land in the west of Scotland. There
must have been an inner sheltered estuary and an outer open
and stormy margin. This is shewn as well by the contrasted
state of the shells, much broken in the Arran beds, perfect
and well preserved even in the most delicate species in the
beds of the inner frith and of Clydesdale, as by the absence
of the laminated clay in Arran, and in other places, such
as Lochgilphead, which must have been exposed in the
state of the surface imagined by us. Clydesdale, in fact,
formed a gulf sheltered by the Lennox hills and coast
range of Renfrew, and was connected to the outer waters
by a strait below Erskine, and a long narrow channel in
the direction of the Dairy valley. Bute and the con-
nected isles would form, for a long period, a low archipel-
ago, sheltering the inner frith; Bute itself consisting of four
islets. Cantire was crossed by several channels, admitting
the ocean — one of them debouching upon Lochgilphead, in

z

333 GEOLOGY OF BUTE.

the line of the Crinan Canal. But then, as now, the moun-
tains of Arran and the Gowal district formed a great outer
barrier against the access of the ocean. Thus, throughout
all the inner waters of the frith of that time, conditions
existed favourable to the development of beds of fine sedi-
ment which the stormy waters and tidal currents, sweeping
the south coast of Arran, would not allow to settle down
over the surface of the boulder-clay when the land was
sinking. The origin of this stoueless clay is very difficult to
understand. Its close resemblance to glacial mud must
strike every one who has examined the deposits in the quiet
bays on the lake of Brienz. We have already traced, in
speaking of the Arran drifts, the probable origin of the other
beds; the origin of the laminated clay is now the most
difficult part of the problem. If deposited on the sea
bottom, by currents passing in from the outer ocean under
the frozen surface of the inner frith, the total absence of
fossils is difficult to understand, though the absence of stones
would be expkined ; whereas, if formed from mud drifted off
from the ends of glaciers by currents passing along shore,
or dropped from ice floating about in the unfrozen frith with
a load of earth and stones, the absence of stones from the
clay is equally difficult to explain. The fineness and perfect
lamination of the clay, and the total absence of sand and
gravel, indicate a slow and tranquil deposit, far out from the
shore, in a deep sea with access to the ocean, but either
frozen over or free from floating ice. We cannot suppose
it so deep that no shells could exist, since we know there
is life at all depths, and that an enormous elevation of the
bottom took place before the deposit of the Arctic shell bed
began over it. Difficulties, indeed, beset every supposi-
tion regarding its origin : it cannot yet be positively affirmed
that it is not really a terrestrial and not a marine formation ;
and we can only wait for more extended observation of the
conditions under which it is associated with the other beds,
and a strict definition of its geographical limits. We shall

GEOLOGY OF BUTE. 339

then have grounds for a safe generalisation. Meanwhile we
may note the great significancy of the contrasts we have
stated, in regard to any theory of its origin. It is persistent
in Clydesdale and the sheltered shores of the frith ; and it
exists at Errol, in Perthshire, an inland sheltered situation ;
but it is absent on the shores of Aberdeen; at Lochgilphead,
a situation once exposed, on an open frith ; and is not found
on the exposed shores of Arran, where also the state of the
shells in the bed that takes its place is very different from
that in which they are found in places where the laminated
clay occurs. From this clay to that above it the change is
abrupt everywhere. New conditions were suddenly intro-
duced all over the district. Animal life was rapidly and
richly developed on the sea bottom ; on this pecular horizon
the sediment became less fine, was more rapidly let down,
and it was charged in many places, but unequally, with
stones, dropped no doubt from floating ice. The life period,
as indicated by the fauna of the shell bed, is remarkable,
and proves clearly the prevalence of a severe climate; but
precise data are yet wanting in regard to successive stages in
this period. Yet are there indications of a greater severity
at first, and a gradual amelioration afterwards, during the
progress of which those species made their appearance which
still remain with us, though rarely found; and the conditions
became assimilated to those which now prevail; the place of
the Arctic species being gradually occupied by that assem-
blage which now fills our teeming waters.

THE AOTIQUITIES OF BUTE.

145. THE Antiquities of the Island of Bute, in common
with tbose wldei. are found in moat of the islands in the
British seas, are interesting, not so much from their
extent as Ikiii variety, and the [local peculiarities which

..-:•.- ;>-'.:.-- ;•- :. : - :_ark< '. in :':.-: ; :••::..-- M
remains, which dinner little from those of the same period
tkroagjhoaA fiarope and a. great part of the East, as in the
• : --..,:.:_-:. •_-: :.:,-. I:. :':. - :••-;•••• : :'. • v :- :. -• v

.::_:_• ;•-•.-• ::!.•.•.-.:'.;•-::. .-::.•:--•..!-.. ' • :..- :
wiA in tne western part of Europe, from Icelan d to Corn-
waDL Hmny of these islands, irtiliili il from the eighth
caatamy by the iliiii|ilni of 8t fntrick and St. Coiamba, had
been previonvly oecopked by the noe who erected tbe
ciich% •MmoHsh^ and cnoH Ant atffl odbt In a brief
acoomnt like the present, it woald be out of place to enter
into a detailed deacription of aO the prehistoric objects in
B«^ eapecxaDy as it too often occors that the merest tcace*

and afipeaiancie, therefore, of the
these antiqnitie» of the prehistoric age
wiD only be given here, reserving far the later and historical

14&. The ntoat interesting saomuncnts of this age in the
the atone aide* of Kicgarth and 84. CaLugs, or
t it is nanHod «n Ae Ordnance avap.

JLSTIQl f II JE& OF BCTK. 341

arirtrA, wW* pgifalp rf fiwg

stones, the «Jh*My» between them imauaui il am ijbeesroBm-
ferance being semen, eight, eleven, lumkiau, thirteen, and
twelve yaioX taken euuaguttii«ly- Of these five stones, only

of the other two. The laigest stone, which, has been deft
in two, apfarcmtly after being set op, m nine feet six inches
high, by seven feet six inches broad. One of the otfeer two
stones is •«»««M*i«M«*i fi«« its uwiiiblanee to a rode cross,

the total height being seven feet six indhra, breadth of disc
seven feet six indhes.

The circle at St. Calmags is •"•pl*** as regards the
number of itonM, which, m seven; one of them is placed
behind another, probably at a recent period, m pmition de-
tlroyingthe sjmmfflij of tihe erode. The atones of the code
are much smaller than those at Kingarth,, tike largest being
seven feet six inches hi^h, and some of them not more than
a foot above the ground, A single stone, «*"'c"fl in m field
not fiur from, this, is the sole remnant of a huge circle
destroyed many years ago.

About half a mile fiom the Kimmrth circle are three
standing stoncr, in one straight line, evidently not forming
put of an indosore; they are not remarkabie fir sue,
and are composed of the local tzap-rock.

At the extremity of $L Hunka's Bay, two standing stones
remain, six or seven feet apart; they are also of modfnto
dimensions, ami may be later than those just mentioned.
There is a single stone at Scalpsie Bay, and one or two
otiken throughout the island of no specal inhnsb

Thiilf jeaiA ago, there were the remains in Bate of iMiimm
hilllmla, the greater number of which have now dhsfnammL1*
Of those remaining, the most iui<jgaUag at thu »iliifiiiifcrt
of Dun-na-goil at the southern extremi)^ of the iebnd. It
i an eminence, fifty feet fciga, protected on all i

342 ANTIQUITIES OF BUTE.

except towards the sea, by the steepness of the cliff' of natural
rock. The remains of the vitrified wall are seen in no
place higher than four feet, and they are about three in
thickness. The vitrified rock is the local whiustone, built
as a dry wall, and fused into one mass, by heat applied from
the outside, as the inside presents no trace of the action of
fire. The plan of the fort is rudely elliptical with the longer
axis almost due N. and S.

Our companion, who is busy with his hammer and lens,
desires us to note the complete glazing of the surface in many
places, and the strongly porphyritic aspect given to the rock
by the perfect fusing and reformation of the felspar of the
trap rock.

The tumuli which were formerly frequent in the island
have nearly all given way to modern improvements, and do
not require especial notice.

There are several cairns, one of which recently excavated
by the Marquis of Bute, on Scalpsie Hill, seems to have been
adapted for a fort at a later period, by building a dry stone
wall seven feet thick, and breast high, on its summit.

Traces of two cranoges have been discovered and described,
one on Loch Quien, and the other on Loch Dhu.* Stone
kists with baked clay urns, of a similar description to those
excavated at Tormore, have occasionally been found. There
are a few instances also, of stone celts and arrow heads
being dug up.

Ecclesiastical Remains.

147. The earliest monuments of the Historic period are un-
doubtedly the ruined chapels scattered along the coast, and
bearing for the most part the names of the Irish saints who
founded them in the first instance. Of these, St. Ninian's
is a characteristic example ; it is situated on the bay of that
name, and its dimensions are, fourteen feet six inches by
twenty -four feet interior measurement, having walls three feet
• Vol. iil, Trans. S.A.S.

ANTIQUITIES OF BUTE.

343

thick built of slate, with clay used in place of mortar. In
this respect it is identical with churches of the same period
(eighth century) remaining in Ulster. Its walls at no place
are higher than four feet, no cut-stone is used about it, and
there is no division into nave and chancel.

Ninian, whose name this chapel bears, was by birth a
Briton, and Roman by education. He founded a mon-
astery at Rosuat, in Wigtoushire, which was afterwards
known in Saxon history as Whithorn, and as Candida Casa
in the Latin writings. Here he educated many distinguished
Irish ecclesiastics, one being Finian of Movilla, in the
County of Down, who in his turn became the teacher of
the illustrious Columkille. Ninian died at the monastery
of Cluain-coner, in the County of Kildare.

At St. Calmags, a corruption probably of St. Colman,
several saints having this name, the chapel has disappeared;
but a stone slab with an incised cross of Irish character, and
of tenth century date, still remains.

The cemetery of St. Blane's chapel contains the greatest
number of these interesting memorials. They seem to have
been used down to the present day as headstones, being
placed at the top of comparatively recent graves.

Fig. 42.
Incised Cross, St. Calmags.

Fig. 43.
Stone Cross, St. Blane's.

They are much weather-worn, but still exhibit the charac-
teristic circular foiled head of varying design, with the shaft
buried almost completely in the ground. Most of these
crosses are of small size, the largest being two feet high, and
many of them smaller. There is, however, the shaft of a

344

ANTIQUITIES OF BUTE.

cross, of which the head is broken off, that has been at
least seven feet high, and possibly a large squared stone in
which a mortice one foot four inches deep has been sunk was
the base of the cross. This large stone lies at present about
fifty feet from the shaft. Some fragments of crosses are
preserved at a farm house near the chapel ; they do not differ
essentially from the others.

The chapel of St. Blane, or St. Blaan, an Irish saint who
flourished in the seventh century, is later in date than the
crosses in question, the present nave being built A.D. 1100, it
is said, by Malcolm Canmore. A chancel was added to the
nave at the close of the thirteenth century. The arch divid-
ing the nave from the chancel has been originally the door-
way into the twelfth century church, as there are two stones,
one on each side of the archway, and next the chancel,
morticed to receive a door frame, and the impost of the arch
is cut away for that purpose, so that the door might close
against it tightly.

The archway is five feet two inches wide, and is a
good example of early twelfth century work. The impost
moulding runs round the nave walls, and is carved with
a variety of the nail-head pattern. The nave walls are
faced inside and out with carefully built sandstone ashlar,
in courses of varying heights. The gables over the chancel

arch and the east window are
the only ones standing ; they
have been evidently repaired
lately. The chancel has been
lighted by two lancets in
the east gable, and one in each
side-wall ; a large window has
been inserted on one side,
blocking up the lower part of
the original window, and is of
fifteenth century date. The
piscina, a pointed arch cut out

Fig. 44.

View of Chancel Arch,
St Blanc's.

ANTIQUITIES OF BUTE. 345

of a single stone, and a square headed ambry still re-
main in their original places. The chancel is constructed
of ashlar, but not of such a good description as the earlier
work.

The whole church is built on an artificial mound or cashel
composed of a substructure of stones, on two different levels,
on which earth has been heaped. Close to the chapel is a holy
well, lined with stone, and with a stone weighing more than
a ton placed on the top. This rude masonry seems coeval
with a circular enclosure not far from it, composed of large
blocks of trap, arranged in the form of a low wall, thirty-three
feet in diameter. This was probably a place for meditation, and
used as such by St. Blane and his immediate successors.
There are traces of extensive foundations in the immediate
vicinity, probably of monastic buildings, in connection with
the chapel.

It is highly probable that this church occupies the site of
a small Culdee settlement, of the early part of the seventh
century. Blaan or Blane, whose day is fixed both in Irish
and Scotch calendars on tenth August, was by his father's
side one of the Dalaradians of Ulster. His mother was the
daughter of Aidau, the King of the Scotch Dalriads. Blaan
was the head of a monastic fraternity at Dunblane, similar
to that at lona founded by Columba, to which he also gave
his name, and which acquired such a prescriptive sanctity
that many centuries afterwards, when dioceses were being
defined, here was raised the beautiful cathedral church of
the See of Dunblane, the limits of which were conterminous
with the Earldom of Strathearn.

The only other church on the island of any note is the old
Cathedral Church of St. Mary of Rothesay, beside the Estab-
lished Church. The chancel alone remains ; it is exactly of
the period, and with the same style of detail as the chancel
of St. Blane's, and like it built of sandstone. It has been
lighted by a large traceried window in the east gable, and by
pointed lights in the side walls. The square-headed ambry and

346 ANTIQUITIES OF BUTE.

piscina, with pointed arch and label, still remain. The best
features, however, are two tombs, both about the commence-
ment of the fifteenth century, placed one on each side of the
chancel, and deeply recessed under a depressed ogival arch.
One tomb contains the effigy of a knight in the armour of
the period, said to be of the royal family, from the crests
on the shields. It is placed under the arch mentioned before,
crocketed, and with fiuial composed of a shield supported by
two animals. Below the shelf on which the effigy lies, the
space is panelled with quatrefoils, and in the centre a shield,
with two angels as supporters. The tomb opposite is that
of a lady and child, and seems slightly later in date; the
arch is more depressed, and the crockets are greatly defaced.
It is panelled beneath the effigy with eight figures in com-
partments, one in each, and nearly all female. Both tombs
are much injured from the effects of time and violence,
having been exposed for nearly two centuries to the
elements.

Chapels of about the same date as Rothesay existed for-
merly at St Bride's, on Chapel Hill, only lately destroyed,
and at Kilmichael, where some foundations remain. On
Inchmarnock a chapel of earlier date stood till lately, but
it has also been destroyed.

Military Remains.

148. The castles, of which there are any remains, are Rothe-
say, Kames, Wester Kames, Kilmorie, and Kelspoke. None
of these are extensive, except Rothesay, which, owing to the
fact of its being a royal castle, was always of great import-
ance. Part of it was probably erected by Magnus, king of
Norway, in 1098, and large additions were made at different
periods subsequently. The greater part of the present build-
ings are of late date, having the characteristic features of the
sixteenth and seventeenth centuries.

The plan comprises the usual courtyard, with four large

ANTIQUITIES OF BUTE. 347

circular flanking towers at the angles. Buildings of all
periods and of various kinds, including a chapel, have been
added ; the whole is surrounded by a ditch or moat, now dry.

The main features of this fine ruin are fast becoming
undistinguishable from the abundant growth of ivy rapidly
spreading over the walls. There are some curious objects
exhibited in connection with the castle ; among others an
early tombstone of a Runic type, with the characteristic
interlaced knot- work, and some carved figures introduced.

Kames Castle, formerly the seat of the Bannatynes of
Kames, consists of a single tower, probably early fifteenth
century work, to which a modern house has been added iu
front. The lower floors of the tower are strongly vaulted
in stone, and the walls are of great thickness. Wester
Kames, not far from the last-named castle, consists of a small
tower in ruins, with a bartizan on the angle. It is of late
date, and has never been a place of any importance.

Kilmorie Castle was the residence of the Jamiesons, the
coroners of Bute in the seventeenth century. A small single
round tower, pierced with one loophole, and some fragments
of flank walls, against which is a modern farm house, are
all that now remain.

Domestic Remains.

149. Kelspoke Castle, situated at the south of the island
on a commanding height S. of the limestone quarry, has been
rather a fortified house of late date than what is usually
understood by the term castle. The plan of a good part of
the buildings can still be traced, as the walls are at some
places breast high. There is nothing, however, of a character
to determine the exact date.

There is an old house in Eothesay, built in 1680, and
which the Bute family occupied whilst their mansion at
Mount Stuart was being erected. It is a very good example
of the Scotch town house of the period ; having a porch of

040 ANTIQUITIES OF BUTE.

two stories with corbie-stepped gables, occupying the breadth
of the foot-path.

The New Statistical Account makes no mention of the
discovery of any objects of antique workmanship in Bute, nor
have we noticed such in Mr. Reid's History. We are,
however, informed by Mr. John Young, of the Hunterian
Museum, of an interesting discovery made in 1862 or 1863.
In digging the foundations of a bridge oil the east coast, near
Kilchattan, there were found three bronze swords, two rude
massive gold rings, and two narrow ribbands of gold. The
swords were of a beautiful form, and in fine preservation.
The gold rings appeared to have been hammered out of a
piece of virgin gold, and were not soldered or fastened in any
way; they were as round as a wire, and had no ornamenta-
tion of any kind ; there were only slight marks of hammering.
The ribbands of gold were several inches in length, very thin,
and a little more than a quarter of an inch broad. They
were marked with fine punctures along both edges, but had
no other ornamentation.

349

THE OTHEE CLYDE ISLANDS.

INCHMARNOCK.

150. This small island lies on the western prolongation of
the outer slate band which crosses Bute, and is noted for its
quarries of roofing slate.

GREAT CUMBRAE.

151. This island is three miles long and one and a-half
broad; the surface rises gently towards the middle, and the
height nowhere exceeds 500 feet; the southern shore presents
some low rocky cliffs; on the other sides the ground descends
by grassy slopes and sandstone ledges. Wanting trees and
streams, and without lofty rocks, the island has little variety
of scenery; bxit the views to be had from almost every
part of it are of surpassing magnificence and beauty. Great
Cumbrae corresponds in geological structure with the middle
region of Bute, and consists of red sandstone, having a slight
northerly dip, traversed by numerous trap dikes, invaded
by sheets of the same rock, but not overlaid by trap except
in one spot on the west coast, where a small patch occurs.
The sandstone is generally of a deep red colour; almost the
only exception being found in the small islets forming the
harbour of Millport, which consist of white and gray sand-
stone. Strata of conglomerate structure are occasionally
met with, and the ridges on the west formerly shewed some
thin limestone beds interstratified with the sandstone, but
these have been worked out. No fossils have been met
with ; but there can be little doubt that the sandstone is of

350 OTHER CLYDE ISLANDS.

the same age as that of S. Bute, most probably lower car-
boniferous. The only varieties of trap occurring here are
coarse and fine greenstone, and basalt ; the fine greenstone
has often a porphyritic structure, from the imbedded felspar
crystals, thus forming the variety incorrectly termed trap
porphyry. The chief interest of the island is in its whin
dikes. These alter the strata remarkably : the sandstone
has been fused by them, and reconsolidated into a substance
closely resembling a dark quartz rock; simple induration is
induced at a greater distance from the dike. Many of the
dikes stand out boldly from the adjoining sandstone, which
has been worn away on either side, the amount of wearing
in the ancient tideway, compared with that in the present,
affording a rude measure of the time during which the sea
remained at the higher level. Attention was first called to
these remarkable dikes by Mr. Smith of Jordanhill. The
best example is seen a little to the east of the entrance of
Millport harbour, where a large dike, rising from the sea
level like a huge wall, runs far up along the hill-side, raised
as if by art above the surface of the fields. On the shore of
the mainland opposite, another dike, having the same
direction and apparently a prolongation of this, stands out
in the same manner from the surface of the sandstone. The
similarity of the two masses of rock and their former con-
nection are pointed at in the legend that this strait was once
spanned by an enormous bridge, raised by the hands of mighty
wizards, of which the only portions that remain are these two
ancient abutments.*

Such projecting masses of trap are not confined, however,
to the shore; they occur in the inland parts, where they
are, of course, not so easily accounted for. We can hardly
suppose them due to atmospheric causes, as there are no
streams to carry off such worn materials ; and they are most

• Quart. Jour. Agric., and Trans. High. Soc., No. xliiL; or voL ix.
p. 430.

OTHER CLYDE ISLANDS. 351

probably due to the action of the sea upon the sandstone
during the progress of the last elevation of the land.

A very interesting shelly deposit here has been described
by Rev. H. W. Crosskey and Mr. D. Robertson. It is
situated near Millport, on a terrace 32 feet 9 inches above
sea level, and 200 yards from the shore, near the resi-
dence of the Hon. Mr. Boyle. The section gives (order
descending) —

Earth Mould, . . . 1| feet.

Gravel, 4 ,,

Shell-bearing Sand, . . 14 ,, (not pierced through).

There were collected forty-four species of shells, — one of
polyzoa, four Crustacea, two echinoderniata, and seventeen
of foraminifera. The character of the shells places it among
the glacial beds. For a detailed list, see Tr. Geol. Soc. GL,
vol. iii. p. 116. A remarkable circumstance about the
deposit is its sandy character, — only ten ounces of clay in
seven and a-half pounds of the deposit. It resembles in
this respect the croffcamic bed on the Forth and Clyde
Junction Railway, (see Jour. Geol. Soc.) The gravel con-
tained fragments of trap (local), quartz (local), and schist
(travelled).

LITTLE CUMBRAE.

152. The Lesser Cumbrae is about two miles long and one
mile broad, and in its bolder outlines strongly contrasts with
the sister isle; its highest point has an elevation of about
800 feet. In geological structure and in altitude it cor-
responds with the southern division of Bute; and, like the
Garrochhead district of that island, consists wholly of trap —
sandstone only appearing in one spot on the east shore,
near the old keep. This patch, however, is sufficient to
shew that the foundation of the island is made of the same
sandstone which forms the Great Cumbrae and the ad-
joining mainland. Piled up over the sandstone foundation

352 OTHER CLYDE ISLANDS.

in great successive sheets, these trap rocks give a terraced
and ridgy structure to the island; and, rising south-west,
present steep, abrupt cliffs towards the sea, while they
decline in succession in a N.E. direction, or contrary to the
inclination of the rocks of the Garrochhead. This is no
doubt due to the original inclination of the surface over
which the submarine lava streams were spread out; but that
inclination is most probably due to a great fault in the sand-
stone, on which the sea, working more successfully along a
fracture, opened this main entrance into the proper estuary
of the Clyde.

There is a very interesting variety among the trap rocks
here, and they often approach the form of perfect columns.
They differ from the Garrochhead traps; and are much more
closely related in mineral structure to those of the mainland.

Mr. Smith of Jordanhill has remarked, that in exact
correspondence with the greater hardness of the rock, the
terrace in front of the old sea-cliff is narrower than in Great
Cumbrae; and he finds glacial striae marked upon the surface
of the terrace parallel to the cliff, and extending, still unworn,
beneath the present level of the sea — an observation of the
utmost importance. See our remai'ks in regard to the
terraces at lorsa water-foot (Newer Pliocene GeoL, p. 144;
and Jour. GeoL Soc., 1862).

The small keep, on a peninsula on the eastern shore, is one
of a range of watch-towers erected along the Clyde shores
during the wars of the Edwards.

PLADDA.

\

153. This islet, we have seen already, is attached to Arran
by a whin dike, over most parts of which there is broken
water at low tide. The island consists almost wholly of a
dark-coloured trap rock, the sandstone foundation appearing
only on the north-eastern shore.

OTHER CLYDE ISLANDS. 353

AILSA CRAIG.

154. This majestic rock rises steeply from the sea to the
height of 1114 feet. It is 13 miles S. of Pladda, and 10
miles "VV. of Girvan, and has an elliptic base 3300 feet by
2200. Its form is nearly that of a right cone, but it is
somewhat more elongated from north to south than in the
opposite direction. This is distinctly seen on climbing to
the summit, which presents a flattened ridge in the former
direction, and is comparatively narrow from east to west.
In shelter from the prevailing winds, a high bank of shingle
has accumulated in the course of ages against the originally
steep face on the east side, and is now perfectly stable.
From the top of this bank the rock rises in broken terraces,
\ip which the ascent is not difficult; on other sides it is
inaccessible. An old stone keep, with vaulted apartments-
one of the line of watch-towers — is perched upon a terrace
on this side, at about one-fifth of the height. With the
exception of the shingle bank, the isle consists wholly of a
reddish-coloured, close-grained syenitic greenstone, quite un-
like any other of the Clyde traps. On the Ayrshire coast,
opposite to Ailsa, a similar syenite is said to occur amid
strata of silurian age. The rock consists of red felspar and
hornblende, and occasional grains of quartz, and is inter-
mediate between a true syenitic or hornblendic granite and
syenitic greenstone. It is thus of igneous origin and a
member of the trap family, though not a basalt, as often
stated. The rock, however, affects the columnar form, and
the precipices on three sides exhibit magnificent ranges of
pillars, which reach the height of fully 400 feet ; but these
are much less perfect than in basalt, the angles are less
sharp, and the concavo-convex joints are wanting, the
divisions being simple seams. The upper 700 feet consist
of amorphous rock; but there is no perceptible difference
between the mineral structure of this and the columnar
part. The precipices are traversed by numerous basaltic

•2 A

354 OTHER CLYDE ISLANDS.

dikes, which simply separate the rock, without inducing
any change of structure upon it. Mr. William N. JVIac-
Cartney, of Glasgow, discovered on Ailsa several years
ago the rare plant Lavatera arborea, or tree mallow. It
grows high on the southern cliffs, and reaches the unusual
height of 6 feet. He has carefully collected all the plants
during two visits — our own was too hurried to admit of
this — and finds 109 species of flowering plants, of which 14
are grasses. The cryptogamic flora is unusually abundant
for a space so limited — the peculiar situation, but still more,
perhaps, the terraced character of the sheltered side, favouring
a prolific vegetation.

FINIS.

COLLINS' SERIES OF FIRST-CLASS ATLASES.

Carefully Constructed and Engraved from the Latest Authorities, and Beautifully
Printed in Colours.

" A complete series of Atlases, chiefly for schools, suited to various stages of
advancement in geographical knowledge, which are highly deserving of com-
mendatory notice. Like the Dictionaries lately published by the same firm,
these Atlases may safely be pronounced National. For cheapness, correctness,
and beauty of style, indeed, we have not seen yet anything to compare with
them." — Morning Journal.

" A series of Atlases for educational purposes, which unite the rare advan-
tages of singular merit and unusual cheapness. The series really supplies a
want, for it notes the various changes rendered necessary within the last few
years, both by political events and by the progress of geographical discovery.
The maps are beautifully engraved, tastefully coloured, and, so far as we have
been able to examine, leave nothing to be desired on the score of accuracy." —
Citizen.

MODERN GEOGRAPHY— Crown Series.

MY FIRST ATLAS, consisting of 12 Maps, 9 inches by 7£ inches, *. </.
folded 8vo, in neat Wrapper, . . . . . .06

THE PRIMARY ATLAS, consisting of 16 Maps, 9 inches by 7J inches,

4to, stiff Wrapper, .06

THE POCKET ATLAS, consisting of 16 Maps, same as aoove, folded

in 8vo, and mounted on Guards, cloth lettered, . . .10

THE YOUNG CHILD'S, OR JUNIOR ATLAS, consisting of 16 Maps,
same as above, 4to, with 16 pp. of Questions on the Maps, in neat
Wrapper, . . , . . . . . .10

THE SCHOOL BOARD ATLAS, consisting of 24 Maps, Crown 4to,

cloth lettered, . . . . . . . . .10

THE PROGRESSIVE ATLAS, consisting of 32 Maps, 9 inches by 7J

inches, 4to, cloth lettered, . . . . . . .20

THE CROWN ATLAS, consisting of 32 Maps, as above, on Guards,

with Index, 8vo, cloth lettered, . . . . . .26

THE NATIONAL ATLAS, consisting of 32 Maps, same as above, 4to,

with a copious Index, cloth lettered, . . . . .26

MODERN GEOGRAPHY — Imperial 4to Series.

s. d.
THE SELECTED ATLAS, consisting of 16 Maps, Imperial 4to, stiff

cover, . . . . . . . . . .16

THE PORTABLE ATLAS, consisting of 16 Maps, same as above,

folded Imperial 8vo, cloth lettered, . . . . .20

THE ADVANCED ATLAS, consisting of 32 Maps, Imperial 4to, cloth

lettered, . . . . . . . . .36

THE ACADEMIC ATLAS, consisting of 32 Maps, Imperial 4to, with

a Copious Index, cloth lettered, . . . . . .50

THE STUDENT'S ATLAS, consisting of 32 Maps, and 6 Ancient

Maps, with a Copious Index, Imperial 8vo, cloth lettered, . 60

THE COLLEGIATE ATLAS, consisting of 32 Modern and 18 Ancient
and Historical Maps, mounted on Guards, with a Copious Index,
Imperial 8vo, cloth lettered, ..... 76

PHYSICAL GEOGRAPHY — Imperial Series.

THE PORTABLE ATLAS OF PHYSICAL GEOGRAPHY, 20 Maps, s. d.
mounted on Guards, imperial 8vo, cloth, . . . .36

THE STUDENT'S ATLAS OP PHYSICAL GEOGRAPHY, 20 Maps,
Guards. With Letterpress Description and Wood Engravings. By
James Bryce, LL.D., F.R.G.S. Imperial 8vo, cloth, . . .60

1. Hemispheres — showing propor-

tions of Land and Water.

2. Europe— showing Mountains and

Rivers, High and Low Lands.

3. Asia — showing as above.

4. Africa — showing as above.

6. North America — showing as

above.
C. South America— showingas above.

7. British Isles — showing as above.

8. Various Sections across the Con-

tinents.

9. Various Sections in the Oceans;

Suez Canal, &c.

10. Maps — showing the Mean Tem-

perature of the Air, March to
August

11. Maps — showing the Mean Tem-

perature of Air, September to
February.

12. Map— showing the Distribution of

Constant, Periodical, and Vari-
able Winds.

13. Map— showing the proportionate

Distribution of Rain, &c.
li. Map— illustrating Earthquakes and
the Distribution of Volcanoes.

15. Map— showing the Ocean Currents

and Basins, and principal River
Systems.

16. Map — showing the Geographical

Distribution of Birds and Fishes.

17. Map— showing the Geographical

Distribution of Quadrupeds anil
Reptiles.

18. Map— showing the Geographical

Distribution of Plants.

19. Map— showing the Distribution of

mankind according to Races.

20. Geological Map, British Itles.

HISTORICAL GEOGRAPHY.

THE POCKET ATLAS OF HISTORICAL GEOGRAPHY, 16 Maps, ».
mounted on Guards, Imperial 16mo, cloth, . . . .1

THE CROWN ATLAS OF H^TORICAL GEOGRAPHY, 16 Maps,
Description by Wm. F. Collier, LL.D., Impena

with Letterpress Des
16mo, cloth,

and

1. Roman Empire, Eastern

Western, 4th Century.

2. Europe, Oth Century, showing Set-

tlements of the Barbarian Tribes.

3. Europe. 9th Century, Empire of

Charlemagne.

4. Europe, 10th Century, Rise of the

German Empire.

5. Europe, 12th Century, Time of the

Crusaders.

6. Europe, Eve of Reformation.

7. Germany, KHh Century, Thirty

Years1' War.

2 6

Europe, 17th and 18th Centuries.

Kurope at the Peace of 1815.

Europe in 1870.

India, Rise of British Empire.

World, showing Voyages of Dis-
covery.

Britain under the Romans.

I'ritain under the Saxons.

Britain after Accession of William
the Conqueror.

France and Belgium, illustrating
I'.ntish History.

CLASSICAL GEOGRAPHY.

COLLINS1 CROWN ATLAS OF CLASSICAL GEOGRAPHY, ODD-
slating of 17 M.-ipx, Imperial 16mo, cloth, with Descriptive Letter-
press, by Dr. 1. SI-II.MII/, and Copious Index,

1. Orbls Verteribus Notus.

2. Egypt.

8. Macedonia, Thraria, &p.

4, 6. Kmiiati Kmpirc, with Plan of

Rome.

6, 7. Grsecia, with Plan of Athens.
H. Empire of Alexander the •

9. Italia (pars Septeutriouulis).

». <l

2 0

10. Italia (pars Merrdonalis).

11. Armenia, Mesopotamia, &c.

12. Asia M.nor.

lia.

14. llispania.

15. Germania, &c.

16. Palestine (temp. Chrin(i).

17. Roman Britain (BritianiuJ.